[Senate Hearing 117-769]
[From the U.S. Government Publishing Office]
S. Hrg. 117-769
PROTECTING KIDS ONLINE: TESTIMONY FROM
FACEBOOK WHISTLEBLOWER
=======================================================================
HEARING
BEFORE THE
SUBCOMMITTEE ON CONSUMER PROTECTION,
PRODUCT SAFETY, AND DATA SECURITY
OF THE
COMMITTEE ON COMMERCE,
SCIENCE, AND TRANSPORTATION
UNITED STATES SENATE
ONE HUNDRED SEVENTEENTH CONGRESS
FIRST SESSION
__________
OCTOBER 5, 2021
__________
Printed for the use of the Committee on Commerce, Science, and
Transportation
[GRAPHIC NOT AVAILABLE IN TIFF FORMAT]
Available online: http://www.govinfo.gov
__________
U.S. GOVERNMENT PUBLISHING OFFICE
54-110 PDF WASHINGTON : 2023
-----------------------------------------------------------------------------------
SENATE COMMITTEE ON COMMERCE, SCIENCE, AND TRANSPORTATION
ONE HUNDRED SEVENTEENTH CONGRESS
FIRST SESSION
MARIA CANTWELL, Washington, Chair
AMY KLOBUCHAR, Minnesota ROGER WICKER, Mississippi, Ranking
RICHARD BLUMENTHAL, Connecticut JOHN THUNE, South Dakota
BRIAN SCHATZ, Hawaii ROY BLUNT, Missouri
EDWARD MARKEY, Massachusetts TED CRUZ, Texas
GARY PETERS, Michigan DEB FISCHER, Nebraska
TAMMY BALDWIN, Wisconsin JERRY MORAN, Kansas
TAMMY DUCKWORTH, Illinois DAN SULLIVAN, Alaska
JON TESTER, Montana MARSHA BLACKBURN, Tennessee
KYRSTEN SINEMA, Arizona TODD YOUNG, Indiana
JACKY ROSEN, Nevada MIKE LEE, Utah
BEN RAY LUJAN, New Mexico RON JOHNSON, Wisconsin
JOHN HICKENLOOPER, Colorado SHELLEY MOORE CAPITO, West
RAPHAEL WARNOCK, Georgia Virginia
RICK SCOTT, Florida
CYNTHIA LUMMIS, Wyoming
Melissa Porter, Deputy Staff Director
George Greenwell, Policy Coordinator and Security Manager
John Keast, Republican Staff Director
Crystal Tully, Republican Deputy Staff Director
Steven Wall, General Counsel
------
SUBCOMMITTEE ON CONSUMER PROTECTION, PRODUCT SAFETY,
AND DATA SECURITY
RICHARD BLUMENTHAL, Connecticut, MARSHA BLACKBURN, Tennessee,
Chair Ranking
AMY KLOBUCHAR, Minnesota JOHN THUNE, South Dakota
BRIAN SCHATZ, Hawaii ROY BLUNT, Missouri
EDWARD MARKEY, Massachusetts JERRY MORAN, Kansas
TAMMY BALDWIN, Wisconsin MIKE LEE, Utah
BEN RAY LUJAN, New Mexico TODD YOUNG, Indiana
C O N T E N T S
----------
Page
Hearing held on October 5, 2021.................................. 1
Statement of Senator Blumenthal.................................. 1
Letter dated October 4, 2021 to Chairman Blumenthal and
Ranking Member Blackburn from the Natioal Association of
Attorneys General.......................................... 4
Prepared statement from Noor Soomro, National Youth Advisory
Board Member, Sandy Hook Promise Students Against Violence
Everywhere (SAVE) Promise Club............................. 9
Letter dated October 6, 2021 to Hon. Richard Blumenthal and
Hon. Marsha Blackburn from Philip J. Weiser, Colorado
Attorney General and Josh Shapiro, Pennsylvania Attorney
General.................................................... 12
Statement of Senator Blackburn................................... 14
Statement of Senator Wicker...................................... 16
Statement of Senator Klobuchar................................... 25
Statement of Senator Thune....................................... 27
Statement of Senator Schatz...................................... 29
Statement of Senator Moran....................................... 31
Statement of Senator Cantwell.................................... 33
Statement of Senator Lee......................................... 36
Statement of Senator Markey...................................... 38
Statement of Senator Lujan....................................... 40
Statement of Senator Hickenlooper................................ 43
Statement of Senator Cruz........................................ 44
Statement of Senator Lummis...................................... 47
Statement of Senator Sullivan.................................... 48
Statement of Senator Scott....................................... 51
Statement of Senator Young....................................... 56
Witnesses
Frances Haugen, Facebook Whistleblower........................... 17
Prepared statement........................................... 19
Appendix
Response to written questions submitted to Frances Haugen by:
Hon. Roy Blunt............................................... 67
Hon. John Thune.............................................. 68
Hon. Marsha Blackburn........................................ 72
PROTECTING KIDS ONLINE: TESTIMONY FROM A FACEBOOK WHISTLEBLOWER
----------
TUESDAY, OCTOBER 5, 2021
U.S. Senate,
Subcommittee on Consumer Protection, Product
Safety, and Data Security,
Committee on Commerce, Science, and Transportation,
Washington, DC.
The Subcommittee met, pursuant to notice, at 10 a.m. in
room SR-253, Russell Senate Office Building, Hon. Richard
Blumenthal, Chairman of the Subcommittee, presiding.
Present: Senators Blumenthal [presiding], Cantwell,
Klobuchar, Schatz, Markey, Lujan, Hickenlooper, Blackburn,
Wicker, Thune, Cruz, Moran, Sullivan, Young, Lee, Scott, and
Lummis.
OPENING STATEMENT OF HON. RICHARD BLUMENTHAL,
U.S. SENATOR FROM CONNECTICUT
Senator Blumenthal. The meeting and hearing of the
Subcommittee on Consumer Protection of the Commerce Committee
will come to order.
I am very pleased to welcome my colleagues and I want to
thank Ranking Member Senator Blackburn for her cooperation and
collaboration. We have been working very closely, and the
Ranking Member who is here, Senator Wicker, as well as our
Chairwoman, Maria Cantwell, Senator Cantwell, I am sure, will
be here shortly.
Most importantly, I would like to thank our witness,
Frances Haugen, for being here and the two counsel who are
representing her today and I want to give you my heartfelt
gratitude for your courage and strength in coming forward as
you have done, standing up to one of the most powerful
implacable corporate giants in the history of the world without
any exaggeration.
You have a compelling credible voice which we have heard
already, but you are not here alone. You are armed with
documents and evidence and you speak volumes as they do about
how Facebook has put profits ahead of people.
Among other revelations, the information that you have
provided to Congress is powerful proof that Facebook knew its
products were harming teenagers. Facebook exploited teens using
powerful algorithms that amplified their insecurities and
abuses through what it found was an addict's narrative.
There is a question, which I hope you will discuss, as to
whether there is such a thing as a safe algorithm. Facebook
sought teens creating secret accounts that are often hidden
from their parents as unique value proposition, in their words,
a ``unique value proposition,'' a way to drive up numbers for
advertisers and shareholders at the expense of safety, and it
doubled down on targeting children, pushing products on pre-
teens, not just teens but pre-teens, that it knows are harmful
to our kids' mental health and well-being.
Instead of telling parents Facebook concealed the facts. It
sought to stonewall and block this information from becoming
public, including to this committee when Senator Blackburn and
I specifically asked the company, and still even now, as of
just last Thursday when a Facebook witness came before this
committee, it has refused disclosure or even to tell us when it
might decide whether to disclose additional documents, and they
have continued their tactics even after they knew the
destruction it caused.
It is not just that they made money from these practices
but they continued to profit from them. Their profit was more
important than the pain that they caused.
Last Thursday the message from Ms. Antigone Davis,
Facebook's Global Head of Safety, was simple, ``This research
is not a bombshell,'' and she repeated the line, ``not a
bombshell.''
Well, this research is the very definition of a bombshell.
Facebook and Big Tech are facing a big debacle moment, a moment
of reckoning. The parallel is striking. I sued Big Tobacco as
Connecticut's Attorney General. I helped to lead the states in
that legal action and I remember very, very well the moment in
the course of our litigation when we learned of those files
that showed not only that Big Tobacco knew that its product
caused cancer but that they had done the research, they
concealed the files, and now we knew and the world knew, and
Big Tech now faces that Big Tobacco jaw-dropping moment of
truth.
It is documented proof that Facebook knows its products can
be addictive and toxic to children and it is not just that they
made money. Again, it is that they valued their profit more
than the pain that they caused to children and their families.
The damage to self-interest and self-worth inflicted by
Facebook today will haunt a generation. Feelings of inadequacy
and insecurity, rejection, and self-hatred will impact this
generation for years to come. Our children are the ones who are
victims. Teens today looking at themselves in the mirror feel
doubt and insecurity.
Mark Zuckerberg ought to be looking at himself in the
mirror today and yet rather than taking responsibility and
showing leadership, Mr. Zuckerberg is going sailing.
His new modus operandi: no apologies, no admission, no
action, nothing to see here. Mark Zuckerberg, you need to come
before this committee. You need to explain to Frances Haugen,
to us, to the world, and to the parents of America what you are
doing and why you did it.
Instagram's business model is pretty straightforward, more
eyeballs, more dollars. Everything Facebook does is to add more
users and keep them on their apps for longer. In order to hook
us, Instagram uses our private information to precisely target
us with content and recommendations, assessing that what will
provoke a reaction will keep us scrolling.
Far too often these recommendations encourage our most
destructive and dangerous behaviors. As we showed on Thursday,
we created a fake account, my office and I did as a team,
interested in extreme dieting and eating disorders. Instagram
latched on to that teenager's initial insecurities. It then
pushed more content and recommendations glorifying eating
disorders.
That's how Instagram's algorithms can push teens into
darker and darker places. Facebook's own researchers called it
Instagram's ``perfect storm,'' exacerbating downward spirals.
Facebook, as you have put it, Ms. Haugen, so powerfully,
maximizes profits and ignores pain.
Facebook's failure to acknowledge and to act makes it
morally bankrupt. Again and again, Facebook rejected reforms
recommended by its own researchers. Last week Ms. Davis said,
``We're looking at'' no specific plans, no commitments, only
vague platitudes.
These documents that you have revealed provided this
company with a blueprint for providing specific recommendations
that could have made Facebook and Instagram safe for the
company repeatedly ignored those recommendations from its own
researchers that would have made Facebook and Instagram safer.
Facebook researchers have suggested changing their
recommendations to stop promoting accounts known to encourage
dangerous body comparison.
Instead of making meaningful changes, Facebook simply pays
lip service and if they won't act and if Big Tech won't act,
Congress has to intervene. Privacy protection is long overdue.
Senator Markey and I have introduced The KIDS Act, which
would ban addictive tactics that Facebook uses to exploit
children. Parents deserve better tools to protect their
children.
I am also a firm supporter of reforming Section 230. We
should consider narrowing the sweeping immunity when platforms'
algorithms amplify illegal conduct. You've commented on this in
your testimony and perhaps you'll expand on it.
We have also heard compelling recommendations about
requiring disclosures of research and independent reviews of
these platforms' algorithms and I plan to pursue these ideas.
The Securities and Exchange Commission should investigate
your contentions and claims, Ms. Haugen, and so should the
Federal Trade Commission. Facebook appears to have misled the
public and investors and if that's correct, it ought to face
real penalties as a result of that misleading and deceptive
misrepresentation.
I want to thank all my colleagues who are here today
because what we have is a bipartisan congressional roadmap for
reform that will safeguard and protect children from Big Tech.
That will be a focus of our subcommittee moving forward, and it
will continue to be bipartisan.
And finally, I'll just end on this note. In the past weeks
and days, parents have contacted me with their stories,
heartbreaking and spine-chilling stories about children pushed
into eating disorders, bullying online, self-injury of the most
disturbing kind, and sometimes even taking their lives because
of social media.
Parents are holding Facebook accountable because of your
bravery, Ms. Haugen, and we need to hold accountable Facebook
and all Big Tech, as well.
Again, my thanks to you. I am going to enter into the
record a letter from 52 State Attorneys General and from two
members of the Youth Advisory Board of Sandy Hook Promise, as
long as there's no objection.
[The information referred to follows:]
Senator Blumenthal. And I will now turn to the Ranking
Member Senator Blackburn.
STATEMENT OF HON. MARSHA BLACKBURN,
U.S. SENATOR FROM TENNESSEE
Senator Blackburn. Thank you, Mr. Chairman, and thank you
for entering that letter in the record that we have from our
States Attorneys General.
Good morning to everyone. It is nice to see people in this
hearing room and to be here for the hearing today. Ms. Haugen,
we thank you for your appearance before us today and for giving
the opportunity, not only for Congress, but for the American
people to hear from you in this setting and we appreciate that.
Mr. Chairman, I think also thanks to you and your staff
that have worked with our team to make certain that we had this
hearing and this opportunity today so that we can get more
insight into what Facebook is actually doing as they invade the
privacy not only of adults but of children and look at the ways
that they are in violation of the Children's Online Privacy
Protection Act, which is Federal law, and looking at how they
are evading that law and working around it.
As the Chairman said, privacy and online privacy, passing a
Federal privacy standard has been long in the works. I filed my
first privacy bill when I was in the House back in 2012 and I
think that it will be this Congress and this subcommittee that
is going to lead the way to online privacy, data security,
Section 230 reforms, and, of course, Senator Klobuchar always
wants to talk about antitrust, and I have to give a nod.
Senator Markey is down there.
When we were in the House, we were probably two of the only
ones who were talking about the need to have a Federal privacy
standard. Now as the Chairman mentioned, last week we heard
from Ms. Davis, who heads Global Safety for Facebook, and it
was surprising to us that what she tried to do was to minimize
the information that was in these documents, to minimize the
research and to minimize the knowledge that Facebook had.
At one point I even reminded her the research was not third
party research, the research was theirs, Facebook's internal
research. So they knew what they were doing. They knew where
the violations were, and they know they are guilty. They know
this. Their research tells them this.
Last week in advance of our hearing, Facebook released two
studies and said that the Wall Street Journal was all wrong,
that they had just gotten it wrong, as if the Wall Street
Journal did not know how to read these documents and how to
work through this research.
Having seen the data that you've presented and the other
studies that Facebook did not publicly share, I feel pretty
confident that it's Facebook who has done the misrepresenting
to this committee.
Here are some of the numbers that Facebook chose not to
share and, Mr. Chairman, I think it's important that we look at
these as we talk about the setting for this hearing, what we
learned last week, what you and I've been learning over the
past 3 years about Big Tech and Facebook, and here you go.
Sixty-six percent of teen girls on Instagram and 40 percent
of teen boys experience negative social comparisons. This is
Facebook's research. Fifty-two percent of teen girls who
experience negative social comparison on Instagram said it was
caused by images related to beauty. Social comparison is worse
on Instagram because it is perceived as real life but based on
celebrity standards. Social comparison mimics the grief cycle
and includes a downward emotional spiral encompassing the range
of emotions, from jealousy to self-proclaimed body dysmorphia.
Facebook addiction, which Facebook calls conveniently calls
problematic use, is most severe in teens peaking at age 14.
Here's what else we know. Facebook is not interested in
making significant changes to improve kids' safety on their
platforms, at least not when that would result in losing
eyeballs on posts or decreasing their ad revenues. In fact,
Facebook is running sacred as they know that in their own
words, young adults are less active and less engaged on
Facebook and that they are running out of teens to add to
Instagram.
So teens are looking at other platforms, like TikTok, and
Facebook is only making those changes that add to its user
numbers and ultimately its profits. Follow the money.
So what are these changes? Allowing users to create
multiple accounts that Facebook does not delete and encouraging
teens to create second accounts they can hide from their
parents. They are also studying younger and younger children,
as young as eight, so that they can market to them, and while
Ms. Davis says that kids below 13 are not allowed on Facebook
or Instagram, we know that they are because she told us that
they recently had deleted 600,000 accounts from children under
age 13.
So how do you get that many underage accounts if you aren't
turning a blind eye to them in the first place? And then in
order to try to clean it up, you go to delete it and then you
say, oh, by the way, we just in the last month deleted 600,000
underage accounts.
And speaking of turning a blind eye, Facebook turns a blind
eye to user privacy. News broke yesterday that the private data
of over 1.5 billion, that's right, 1.5 billion Facebook users
are being sold on a hacking forum. That's its biggest data
breach to date.
Examples like this underscore my strong concerns about
Facebook collecting the data of kids and teens and what they
are doing with it.
Facebook also turns a blind eye toward blatant human
exploitation taking place on its platform, trafficking, forced
labor, cartels, the worst possible things one can imagine.
Big Tech companies have gotten away with abusing consumers
for too long. It is clear that Facebook prioritizes profit over
the well-being of children and all users.
So as a mother and a grandmother, this is an issue that is
of particular concern to me. So we thank you for being here
today, Ms. Haugen, and we look forward to getting to the truth
about what Facebook is doing with users' data and how they are
abusing their privacy and how they show a lack of respect for
the individuals that are on their network. We look forward to
the testimony.
Thank you, Mr. Chairman.
Senator Blumenthal. Thank you, Senator Blackburn.
I don't know whether Ranking Member Wicker would like to
make a statement.
STATEMENT OF HON. ROGER WICKER,
U.S. SENATOR FROM MISSISSIPPI
Senator Wicker. If you don't mind, thank you, Chairman
Blumenthal, and I will just take a moment or two, and I do
appreciate being able to speak as Ranking Member of the Full
Committee.
Ms. Haugen, this is a subcommittee hearing. You see some
vacant seats. This is pretty good attendance for a
subcommittee. There are also a lot of things going on. So
people will be coming and going, but I'm willing to predict
that this will have almost 100 percent attendance by members of
the Subcommittee because of the importance of this subject
matter.
So thanks for coming forward to share concerns about
Facebook's business practices, particularly with respect to
children and teens, and, of course, that is the main topic of
the title of our hearing today, Protecting Kids Online.
The recent revelations about Facebook's mental health
effects on children and its plan to target younger audiences
are indeed disturbing, and I think you're going to see a lot of
bipartisan concern about this today and in future hearings.
They show how urgent it is for Congress to act against
powerful tech companies on behalf of children and the broader
public, and I say powerful tech companies. They are possessive
of immense, immense power. Their product is addictive and
people on both sides of this dais are concerned about this.
I talked to an opinion-maker just down the hall a few
moments before this hearing and this person said, ``The tech
gods have been demystified now,'' and I think this hearing
today, Mr. Chair, is a part of the process of demystifying Big
Tech.
The children of America are hooked on their product. It is
often destructive and harmful and there is a cynical knowledge
on behalf of the leadership of these Big Tech companies that
that is true.
Ms. Haugen, I hope you will have a chance to talk about
your work experience at Facebook and perhaps compare it to
other social media companies.
I also look forward to hearing your thoughts on how this
committee and how this Congress can ensure greater
accountability and transparency, especially with regard to
children.
So thank you, Mr. Chairman, and thank you, Ms. Haugen, for
being here today.
Senator Blumenthal. Thank you, Senator Wicker.
Our witness this morning is Frances Haugen. She was the
Lead Product Manager on Facebook's Civic Misinformation Team.
She holds a degree in Electrical and Computer Engineering from
Olin College and an MBA from Harvard.
She made the courageous decision, as all of us here and
many others around the world know, to leave Facebook and reveal
the terrible truths about the company. She learned during her
tenure there and I think we are all in agreement here in
expressing our gratitude and our admiration for your bravery in
coming forward.
Thank you, Ms. Haugen. Please proceed.
STATEMENT OF FRANCES HAUGEN, FACEBOOK WHISTLEBLOWER
Ms. Haugen. Good afternoon, Chairman Blumenthal, Ranking
Member Blackburn, and Members of the Subcommittee. Thank you
for the opportunity to appear before you.
My name is Frances Haugen. I used to work at Facebook. I
joined Facebook because I think Facebook has the potential to
bring out the best in us, but I'm here today because I believe
Facebook's products harm children, stoke division, and weaken
our democracy.
The company's leadership knows how to make Facebook and
Instagram safer but won't make the necessary changes because
they have put their astronomical profits before people.
Congressional action is needed. They won't solve this
crisis without your help.
Yesterday we saw Facebook get taken off the internet. I
don't know why it went down, but I know that for more than 5
hours Facebook wasn't used to deepen divides, destabilize
democracies, and make young girls and women feel bad about
their bodies.
It also means that millions of small businesses weren't
able to reach potential customers and countless photos of new
babies weren't joyously celebrated by family and friends around
the world.
I believe in the potential of Facebook. We can have social
media we enjoy, that connects us without tearing apart our
democracy, putting our children in danger, and sowing ethnic
violence around the world. We can do better.
I have worked as a product manager at large tech companies
since 2006, including Google, Pinterest, Yelp, and Facebook. My
job has largely focused on algorithmic products like Google+
Search and recommendation systems like the one that powers the
Facebook News Feed.
Having worked on four different types of social networks, I
understand how complex and nuanced these problems are. However,
the choices being made inside of Facebook are disastrous for
our children, for our public safety, for our privacy, and for
our democracy, and that is why we must demand Facebook make
changes.
During my time at Facebook, first working as the Lead
Product Manager for Civic Misinformation and later on
Counterespionage, I saw Facebook repeatedly encounter conflicts
between its own profits and our safety. Facebook consistently
resolved these conflicts in favor of its own profits. The
result has been more division, more harm, more lies, more
threats, and more combat. In some cases, this dangerous online
talk has led to actual violence that harms and even kills
people.
This is not simply a matter of certain social media users
being angry or unstable or that one side being radicalized
against the other. It is about Facebook choosing to grow at all
costs, becoming an almost trillion dollar company by buying its
profits with our safety.
During my time at Facebook, I came to realize the
devastating truth. Almost no one outside of Facebook knows what
happens inside of Facebook. The company intentionally hides
vital information from the public, from the U.S. Government,
and from governments around the world.
The documents I have provided to Congress prove that
Facebook has repeatedly misled the public about what its own
research reveals about the safety of children, the efficacy of
its artificial intelligence systems, and its role in spreading
divisive and extreme messages.
I came forward because I believe that every human being
deserves the dignity of the truth.
The severity of this crisis demands that we break out of
our previous regulatory frames. Facebook wants to trick you
into thinking that privacy protections or changes to Section
230 alone will be sufficient.
While important, these will not get to the core of the
issue, which is that no one truly understands the destructive
choices made by Facebook, except Facebook.
We can afford nothing less than full transparency. As long
as Facebook is operating in the shadows, hiding its research
from public scrutiny, it is unaccountable. Until the incentives
change, Facebook will not change.
Left alone, Facebook will continue to make choices that go
against the common good, our common good. When we realized Big
Tobacco was hiding the harms it caused, the government took
action. When we figured out cars were safer with seatbelts, the
government took action, and when our government learned that
opioids were taking lives, the government took action. I
implore you to do the same here.
Today, Facebook shapes our perception of the world by
choosing the information we see. Even those who don't use
Facebook are impacted by the majority who do. A company with
such frightening influence over so many people, over their
deepest thoughts, feelings, and behavior needs real oversight.
But Facebook's closed design means it has no real
oversight. Only Facebook knows how it personalizes your feed
for you. At other large tech companies, like Google, any
independent researcher can download from the Internet the
company's search results and write papers about what they find
and they do, but Facebook hides behind walls that keeps
researchers and regulators from understanding the true dynamics
of their system.
Facebook will tell you privacy means they can't give you
data. This is not true. When tobacco companies claimed that
filtered cigarettes were safer for consumers, scientists could
independently invalidate these marketing messages and confirm
that, in fact, they posed a greater threat to human health. The
public cannot do the same with Facebook. We are given no other
option than to take their marketing messages on blind faith.
Not only does the company hide most of its own data, my
disclosure has proved that when Facebook is directly asked
questions as important as how do you impact the health and
safety of our children, they choose to mislead and misdirect.
Facebook has not earned our blind faith. This inability to
see into Facebook's actual systems and confirm that they work
as communicated is like the Department of Transportation
regulating cars but only watching them drive down the highway.
Today, no regulator has a menu of solutions for how to fix
Facebook because Facebook didn't want them to know enough about
what's causing the problems. Otherwise, there wouldn't have
been need for a whistleblower.
How is the public supposed to assess if Facebook is
resolving conflicts of interest in a way that is aligned with
the public good if the public has no visibility into how
Facebook operates? This must change.
Facebook wants you to believe that the problems we're
talking about are unsolvable. They want you to believe in false
choices. They want you to believe you must choose between a
Facebook full of divisive and extreme content or losing one of
the most important values our country was founded upon, free
speech, that you must choose between public oversight of
Facebook's choices and your personal privacy, that to be able
to share fun photos of your kids with old friends, you must
also be inundated with anger-driven virility. They want you to
believe that this is just part of the deal.
I am here today to tell you that's not true. These problems
are solvable. A safer, free speech-respecting, more enjoyable
social media is possible, but there's one thing that I hope
everyone takes away from these disclosures. It is that Facebook
can change but it's clearly not going to do so on its own.
My fear is that without action, divisive and extremist
behaviors we see today are only the beginning. What we saw in
Myanmar and are now seeing in Ethiopia are only the opening
chapters of a story so terrifying no one wants to read the end
of it.
Congress can change the rules that Facebook plays by and
stop the many harms it is now causing. We now know the truth
about Facebook's destructive impact. I really appreciate the
seriousness which the Members of Congress and the Securities
and Exchange Commission are approaching these issues.
I came forward at great personal risk because I believe we
still have time to act but we must act now. I'm asking you, our
elected representatives, to act.
Thank you.
[The prepared statement of Ms. Haugen follows:]
Prepared Statement of Frances Haugen, Facebook Whistleblower
Chairman Blumenthal, Ranking Member Blackburn, and Members of the
Subcommittee. Thank you for the opportunity to appear before you and
for your interest in confronting one of the most urgent threats to the
American people, to our children and our country's well-being, as well
as to people and nations across the globe.
My name is Frances Haugen. I used to work at Facebook and joined
because I think Facebook has the potential to bring out the best in us.
But I am here today because I believe that Facebook's products harm
children, stoke division, weaken our democracy and much more. The
company's leadership knows ways to make Facebook and Instagram safer
and won't make the necessary changes because they have put their
immense profits before people. Congressional action is needed. They
cannot solve this crisis without your help.
I believe that social media has the potential to enrich our lives
and our society. We can have social media we enjoy--one that brings out
the best in humanity. The Internet has enabled people around the world
to receive and share information and ideas in ways never conceived of
before. And while the Internet has the power to connect an increasingly
globalized society, without careful and responsible development, the
Internet can harm as much as it helps.
I have worked as a product manager at large tech companies since
2006, including Google, Pinterest, Yelp and Facebook. My job has
largely focused on algorithmic products like Google+ Search and
recommendation systems like the one that powers the Facebook News Feed.
Working at four major tech companies that operate different types of
social networks, I have been able to compare and contrast how each
company approaches and deals with different challenges. The choices
being made by Facebook's leadership are a huge problem--for children,
for public safety, for democracy--that is why I came forward. And let's
be clear: it doesn't have to be this way. We are here today because of
deliberate choices Facebook has made.
I joined Facebook in 2019 because someone close to me was
radicalized online. I felt compelled to take an active role in creating
a better, less toxic Facebook. During my time at Facebook, first
working as the lead product manager for Civic Misinformation and later
on Counter-Espionage, I saw that Facebook repeatedly encountered
conflicts between its own profits and our safety. Facebook consistently
resolved those conflicts in favor of its own profits. The result has
been a system that amplifies division, extremism, and polarization--and
undermining societies around the world. In some cases, this dangerous
online talk has led to actual violence that harms and even kills
people. In other cases, their profit optimizing machine is generating
self-harm and self-hate--especially for vulnerable groups, like teenage
girls. These problems have been confirmed repeatedly by Facebook's own
internal research.
This is not simply a matter of some social media users being angry
or unstable. Facebook became a $1 trillion company by paying for its
profits with our safety, including the safety of our children. And that
is unacceptable.
I believe what I did was right and necessary for the common good--
but I know Facebook has infinite resources, which it could use to
destroy me. I came forward because I recognized a frightening truth:
almost no one outside of Facebook knows what happens inside Facebook.
The company's leadership keeps vital information from the public, the
U.S. government, its shareholders, and governments around the world.
The documents I have provided prove that Facebook has repeatedly misled
us about what its own research reveals about the safety of children,
its role in spreading hateful and polarizing messages, and so much
more. I appreciate the seriousness with which Members of Congress and
the Securities and Exchange Commission are approaching these issues.
The severity of this crisis demands that we break out of previous
regulatory frames. Tweaks to outdated privacy protections or changes to
Section 230 will not be sufficient. The core of the issue is that no
one can understand Facebook's destructive choices better than Facebook,
because only Facebook gets to look under the hood. A critical starting
point for effective regulation is transparency: full access to data for
research not directed by Facebook. On this foundation, we can build
sensible rules and standards to address consumer harms, illegal
content, data protection, anticompetitive practices, algorithmic
systems and more.
As long as Facebook is operating in the dark, it is accountable to
no one. And it will continue to make choices that go against the common
good. Our common good.
When we realized tobacco companies were hiding the harms it caused,
the government took action. When we figured out cars were safer with
seat belts, the government took action. And today, the government is
taking action against companies that hid evidence on opioids.
I implore you to do the same here.
Right now, Facebook chooses what information billions of people
see, shaping their perception of reality. Even those who don't use
Facebook are impacted by the radicalization of people who do. A company
with control over our deepest thoughts, feelings and behaviors needs
real oversight.
But Facebook's closed design means it has no oversight--even from
its own Oversight Board, which is as blind as the public. Only Facebook
knows how it personalizes your feed for you. It hides behind walls that
keep the eyes of researchers and regulators from understanding the true
dynamics of the system. When the tobacco companies claimed that
filtered cigarettes were safer for consumers, it was possible for
scientists to independently invalidate that marketing message and
confirm that in fact they posed a greater threat to human health.\1\
But today we can't make this kind of independent assessment of
Facebook. We have to just trust what Facebook says is true--and they
have repeatedly proved that they do not deserve our blind faith.
---------------------------------------------------------------------------
\1\ James Hamblin. ``If My Friend Smokes Sometimes, Should the
Cigarettes Have Filters? An honest question.'' The Atlantic. https://
www.theatlantic.com/health/archive/2017/07/cigarette-filters/533379/
---------------------------------------------------------------------------
This inability to see into the actual systems of Facebook and
confirm that Facebook's systems work like they say is like the
Department of Transportation regulating cars by watching them drive
down the highway. Imagine if no regulator could ride in a car, pump up
its wheels, crash test a car, or even know that seat belts could exist.
Facebook's regulators can see some of the problems--but they are
kept blind to what is causing them and thus can't craft specific
solutions. They cannot even access the company's own data on product
safety, much less conduct an independent audit. How is the public
supposed to assess if Facebook is resolving conflicts of interest in a
way that is aligned with the public good if it has no visibility and no
context into how Facebook really operates?
This must change.
Facebook wants you to believe that the problems we're talking about
are unsolvable. They want you to believe in false choices. They want
you to believe you must choose between connecting with those you love
online and your personal privacy. That in order to share fun photos of
your kids with old friends, you must also be inundated with
misinformation. They want you to believe that this is just part of the
deal. I am here to tell you today that's not true. These problems are
solvable. A safer, more enjoyable social media is possible. But if
there is one thing that I hope everyone takes away from these
disclosures it is that Facebook chooses profit over safety every day--
and without action, this will continue.
Congress can change the rules Facebook plays by and stop the harm
it is causing.
I came forward, at great personal risk, because I believe we still
have time to act. But we must act now.
Thank you.
Senator Blumenthal. Thank you, Ms. Haugen. Thank you for
taking that personal risk and we will do anything and
everything to protect and stop any retaliation against you and
any legal action that the company may bring to bear or anyone
else and we've made that, I think, very clear in the course of
these proceedings.
I want to ask you about this idea of disclosure. You have
talked about looking in effect at a car going down the road and
were going to have 5-minute rounds of questions, maybe a second
round if you are willing to do it.
We are here today to look under the hood and that is what
we need to do more. In August, Senator Blackburn and I wrote to
Mark Zuckerberg and we asked him pretty straightforward
questions about how the company works and safeguards for
children and teens on Instagram. Facebook dodged, ducked,
sidetracked, in effect, misled us.
So I'm going to ask you a few straightforward questions to
break down some of what you have said and if you can answer
them yes or no that would be great.
Has Facebook's research, its own research, ever found that
its platforms can have a negative effect on children and teens'
mental health or well-being?
Ms. Haugen. Many of Facebook's internal research reports
indicate that Facebook has a serious negative harm on a
significant portion of teenagers and children.
Senator Blumenthal. And has Facebook ever offered features
that it knew had a negative effect on children's and teens'
mental health?
Ms. Haugen. Facebook knows that its amplification
algorithms, things like engagement-based ranking on Instagram,
can lead children from very innocuous topics, like healthy
recipes, I think all of us could eat a little more healthy, all
the way from just something innocent like healthy recipes to
anorexia-promoting content over a very short period of time.
Senator Blumenthal. And has Facebook ever found, again in
its research, that kids show signs of addiction on Instagram?
Ms. Haugen. Facebook has studied a pattern that they call
``problematic use,'' what we might more commonly call
``addiction.'' It has a very high bar for what it believes it
is. It says you self-identify that you don't have control over
your usage and that it is materially harming your health, your
schoolwork, or your physical health.
Five to 6 percent of 14-year-olds have the self-awareness
to admit both those questions. It is likely that far more than
five to 6 percent of 14-year-olds are addicted to Instagram.
Senator Blumenthal. Last Thursday, my colleagues and I
asked Ms. Davis, who was representing Facebook, about how the
decision would be made whether to pause permanently Instagram
for Kids and she said, ``There's no one person who makes a
decision like that. We think about that collaboratively.'' It
is as though she could not mention Mark Zuckerberg's name.
Isn't he the one who will be making this decision from your
experience in the company?
Ms. Haugen. Mark holds a very unique role in the tech
industry in that he holds over 55 percent of all the voting
shares for Facebook. There are no similarly powerful companies
that are as unilaterally controlled. In the end, the buck stops
with Mark. There is no one currently holding Mark accountable
but himself.
Senator Blumenthal. And Mark Zuckerberg in effect is the
algorithm designer-in-chief, correct?
Ms. Haugen. I received an MBA from Harvard and they
emphasize to us that we are responsible for the organizations
that we build. Mark has built an organization that is very
metrics-driven. It is intended to be flat. There is no
unilateral responsibility. The metrics make the decision.
Unfortunately, that itself is a decision and in the end, if
he is the CEO and the Chairman of Facebook, he is responsible
for those decisions.
Senator Blumenthal. The buck stops with him.
Ms. Haugen. The buck stops with him.
Senator Blumenthal. And speaking of the buck stopping, you
have said that Facebook should declare moral bankruptcy. I
agree.
Ms. Haugen. Yes.
Senator Blumenthal. I think its actions and its failure to
acknowledge its responsibility indicate moral bankruptcy.
Ms. Haugen. There is a cycle occurring inside the company
where Facebook has struggled for a long time to recruit and
retain the number of employees it needs to tackle the large
scope of projects that it has chosen to take on.
Facebook is stuck in a cycle where it struggles to hire.
That causes it to understaff projects, which causes scandals,
which then makes it harder to hire. Part of why Facebook needs
to come out and say we did something wrong, we made some
choices that we regret is the only way we can move forward and
heal Facebook is we first have to admit the truth. The way we
will have reconciliation and we can move forward is by first
being honest and declaring moral bankruptcy.
Senator Blumenthal. Being honest and acknowledging that
Facebook has caused and aggravated a lot of pain simply to make
more money and it has profited off spreading disinformation and
misinformation and sowing hate. Facebook's answers to
Facebook's destructive impact always seems to be more Facebook.
We need more Facebook, which means more pain and more money for
Facebook. Would you agree?
Ms. Haugen. I don't think at any point Facebook set out to
make a destructive platform. I think it is a challenge that
Facebook has set up an organization where the parts of the
organization responsible for growing and expanding the
organization are separate and not regularly cross-pollinated
with the parts of the company that focus on harms that the
company is causing and as a result, regularly integrity
actions, projects that were hard fought by the teams trying to
keep us safe are undone by new growth projects that counteract
those same remedies.
So I do think it is a thing of there are organizational
problems that need oversight and Facebook needs help in order
to move forward to a more healthy place.
Senator Blumenthal. And whether it is teens bullied into
suicidal thoughts or the genocide of ethnic minorities in
Myanmar, or fanning the flames of division within our own
country or in Europe, they are ultimately responsible to the
immorality of the pain that is caused.
Ms. Haugen. Facebook needs to take responsibility for the
consequences of its choices. It needs to be willing to accept
small tradeoffs on profit, and I think just that act of being
able to admit that it's a mixed bag is important, and I think
that what we saw from Antigone last week is an example of the
kind of behavior we need to support Facebook in growing out of
which is instead of just focusing on all the good they do,
admit they have responsibilities to also remedy the harm.
Senator Blumenthal. But Mark Zuckerberg's new policy is no
apologies, no admissions, no acknowledgement, nothing to see
here. We are going to deflect it and go sailing.
I turn to the Ranking Member.
Senator Blackburn. Thank you, Mr. Chairman. Thank you for
your testimony.
I want to stay with Ms. Davis and some of her comments
because I had asked her last week about the underage users and
she had made the comment, I'm going to quote from her
testimony, ``If we find an account of someone who's under 13,
we remove them. In the last 3 months, we removed 600,000
accounts of under 13-year-olds.''
And I have to tell you, it seems to me that there's a
problem if you have 600,000 accounts from children who ought
not to be there in the first place.
So what did Mark Zuckerberg know about Facebook's plans to
bring kids on as new users and advertise to them?
Ms. Haugen. There are reports within Facebook that show
cohort analyses where they examine out what ages do people join
Facebook and Instagram, and based on those cohort analyses, so
Facebook likes to say children lie about their ages to get on
to the platform.
The reality is enough kids tell the truth that you can work
backward to figure out what are approximately the real ages of
anyone who's on the platform. When Facebook does cohort
analyses and looks back retrospectively, it discovers things
like, you know, up to 10 to 15 percent of even 10-year-olds in
a given cohort may be on Facebook or Instagram.
Senator Blackburn. OK. So this is why Adam Mosseri, who's
the CEO of Instagram, would have replied to JoJo Siwa when she
said to him, ``Oh, I've been on Instagram since I was eight,''
he said he didn't want to know that. So it would be for this
reason, correct?
Ms. Haugen. A pattern of behavior that I saw at Facebook
was that often problems were so understaffed that there was
kind of an implicit discouragement from having better detection
systems. So, for example, my last team at Facebook was on the
Counterespionage Team within the Threat Intelligence Org, and
at any given time our team could only handle a third of the
cases that we knew about. We knew that if we built even a basic
detector, we would likely have many more cases.
Senator Blackburn. OK. Then let me ask you this. So you
look at the way that they have the data but they're choosing to
keep that data and advertise from it, right?
Ms. Haugen. Yes.
Senator Blackburn. Sell it to third parties. So what does
Facebook do? You've got these 600,000 accounts that ought not
to be on there and----
Ms. Haugen. Probably more.
Senator Blackburn. Right. But then you delete those
accounts, but what happens to that data? Does Facebook keep
that data? Do they keep it until those children go to age 13
since you're saying they can work backward and figure out the
true age of a user? So what do they do with it? Do they delete
it? Do they store it? Do they keep it? How do they process
that?
Ms. Haugen. My understanding of Facebook's scatter
retention policies--let me be really clear. I didn't work
directly on that--is that they delete--when they delete an
account, they delete all the data I believe within 90 days in
compliance with the GDPR.
With regard to children underage on the platform, Facebook
does substantially more to detect more of those children and
they should have to publish for Congress those processes
because there are lots of subtleties in those things and they
could be much more effective than probably what they're doing
today.
Senator Blackburn. Got it. Now staying with this underage
children since this hearing is all about kids and about online
privacy, I want you to tell me how Facebook is able to do
market research on these children that are underage 13 because
Ms. Davis was really--she didn't deny this last week.
So how are they doing this? Do they bring kids into focus
groups with their parents? How do they get that permission? She
said they got permission from parents. Is there a permission
slip or a form that gets signed and then how do they know which
kids to target?
Ms. Haugen. A bunch to unpack there.
Senator Blackburn. Well, start with maybe how do they
recruit children for focus groups or recruit teenagers?
Ms. Haugen. Most tech companies have systems where they can
analyze the data that is on their servers. So most of the focus
groups I read or that I saw analysis of were around Messenger
Kids which has children on it and those focus groups appear to
be children interacting in person. Often large tech companies
use either sourcing agencies that will go and identify people
who meet certain demographic criteria or they will reach out
directly based on data on the platform.
So, for example, in the case of Messenger Kids, maybe you
would want to study a child that was an active user and one
that was a less active user. You might reach out to some that
came from each population.
Senator Blackburn. And so these are children that are under
age 13?
Ms. Haugen. Yes.
Senator Blackburn. And they know it?
Ms. Haugen. For some of these studies, and I assume they
get permission but I don't work on that.
Senator Blackburn. OK. Well, we're still waiting to get a
copy of that parental consent form that would involve children.
My time is expired. Mr. Chairman, I'll save my other
questions for our second round, if we're able to get those.
Thank you.
Senator Blumenthal. Great. Thank you, Senator Blackburn.
Senator Klobuchar.
STATEMENT OF HON. AMY KLOBUCHAR,
U.S. SENATOR FROM MINNESOTA
Senator Klobuchar. Thank you very much, Mr. Chairman.
Thank you so much, Ms. Haugen, for shedding a light on how
Facebook time and time again has put profit over people.
When their own research found that more than 13 percent of
teen girls say that Instagram made their thoughts of suicide
worse, what did they do? They proposed Instagram for Kids,
which has now been put on pause because of public pressure.
When they found out that their algorithms are fostering
polarization, misinformation, and hate that they allowed 99
percent of their violent content to remain unchecked on their
platform, including lead-up to the January 6th insurrection,
what did they do? They now, as we know, Mark Zuckerberg's going
sailing and saying no apologies.
I think the time has come for action and I think you are
the catalyst for that action. You have said privacy legislation
is not enough. I completely agree with you. I think you know we
have not done anything to update our privacy laws in this
country, our Federal privacy laws, nothing, zilch in any major
way. Why? Because there are lobbyists around every single
corner of this building that have been hired by the tech
industry.
We have done nothing when it comes to making the algorithms
more transparent, allowing for the universe of research that
you referred to. Why? Because Facebook and the other tech
companies are throwing a bunch of money around this town and
people are listening to them.
We have done nothing significantly passed, although we are
in a bipartisan basis working in the Antitrust Subcommittee to
get something done on consolidation which you understand allows
the dominant platforms to control all this, like the bullies in
the neighborhood, buy out the companies that maybe could have
competed with them and added the bells and whistles.
So the time for action is now. So I'll start with something
that I asked Facebook's Head of Safety when she testified
before us last week. I asked her how they estimate the lifetime
value of a user for kids who start using their products before
they turn 13. She evaded the question and said that's not the
way we think about it.
Is that right or is it your experience that Facebook
estimates and puts a value on how much money they get from
users in general? I'll get to kids in a second. Is that a
motivating force for them?
Ms. Haugen. Based on what I saw in terms of allocation of
Integrity's funding, so one of the things disclosed in the Wall
Street Journal was that I believe it's like 87 percent of all
the misinformation spending is spent on English but only about
like 9 percent of the users are English speakers.
It seems that Facebook invests more in users who make the
more money, even though the danger may not be evenly
distributed based on profitability.
Senator Klobuchar. Does it make sense that having a younger
person get hooked on social media at a young age makes them
more profitable over the long term as they have a life ahead of
them?
Ms. Haugen. Facebook's internal documents talk about the
importance of getting younger users, for example, tweens, on to
Instagram, like Instagram Kids, because they need to have--like
they know that children bring their parents online and things
like that and so they understand the value of younger users for
the long-term success of Facebook.
Senator Klobuchar. Facebook reported advertising revenue to
be $51.58 per user----
Ms. Haugen. Oh, wow.
Senator Klobuchar.--last quarter in the U.S. and Canada.
When I asked Ms. Davis how much of that came from Instagram
users under 18, she wouldn't say.
Do you think that teens are profitable for their company?
Ms. Haugen. I would assume so, based on advertising for
things like television. You get substantially higher
advertising rates for customers who don't yet have preferences
or habits and so I'm sure they are some of the more profitable
users on Facebook, but I did not work directly on that.
Senator Klobuchar. Now there's a major issue that's come
out of this, eating disorders. Studies have found that eating
disorders actually have the highest mortality rate of any
mental illness for women, and I led a bill on this with
Senators Capito and Baldwin that we passed into law, and I'm
concerned that these algorithms that they have pushes
outrageous content promoting anorexia and the like.
I know it's personal to you. Do you think that their
algorithms push some of this content to young girls?
Ms. Haugen. Facebook knows that the engagement-based
ranking, the way that they pick the content in Instagram for
young users, for all users, amplifies preferences, and they
have done something called a ``proactive incident response''
where they take things that occur, for example, like can you be
led by the algorithms to anorexia content, and they have
literally recreated that experiment themselves and confirmed
yes, this happens to people. So Facebook knows that they are
leading young users to anorexia content.
Senator Klobuchar. Do you think they are deliberately
designing their product to be addictive beyond even that
content?
Ms. Haugen. Facebook has a long history of having a
successful and very effective growth division where they take
little tiny tweaks and they constantly, constantly, constantly
are trying to optimize it to grow. Those kinds of stickiness
could be construed as things that facilitate addiction.
Senator Klobuchar. Right. The last thing I want to ask is
we've seen this same kind of content in the political world.
You brought up other countries and what's been happening there.
On 60 Minutes you said that Facebook implemented safeguards
to reduce misinformation ahead of the 2020 election but turned
off those safeguards right after the election and you know that
the insurrection occurred January 6.
Do you think that Facebook turned off the safeguards
because they were costing the company money, because it was
reducing profits?
Ms. Haugen. Facebook has been emphasizing a false choice.
They've said the safeguards that were in place before the
election implicated free speech.
The choices that were happening on the platform were really
about how reactive and twitchy was the platform, right, like
how viral was the platform, and Facebook changed those safety
defaults in the run-up to the election because they knew they
were dangerous and because they wanted that growth back. They
wanted the acceleration of the platform back after the
election. They returned to their original defaults and the fact
that they had to break the glass on January 6th and turn them
back on, I think that's deeply problematic.
Senator Klobuchar. Agree. Thank you very much for your
bravery in coming forward.
Senator Blumenthal. Senator Thune.
STATEMENT OF HON. JOHN THUNE,
U.S. SENATOR FROM SOUTH DAKOTA
Senator Thune. Thank you, Mr. Chair, and Ranking Member
Blackburn.
I've been arguing for some time that it is time for
Congress to act, and I think the question is always what is the
correct way to do it? The right way to do it consistent with
our First Amendment right to free speech?
This committee doesn't have jurisdiction over the antitrust
issue. That's the Judiciary Committee. I'm not averse to
looking at the monopolistic nature of Facebook honestly. I
think that's a real issue that needs to be examined and perhaps
addressed, as well.
But at least under this committee's jurisdiction, there are
a couple of things I think we can do, and I have a piece of
legislation and Senators Blackburn and Blumenthal are both co-
sponsors called The Filter Bubble Transparency Act and
essentially what it would do is give users the options to
engage with social media platforms without being manipulated by
these secret formulas that essentially dictate the content that
you see when you open up an app or log on to a website.
We also, I think, need to hold Big Tech accountable by
reforming Section 230 and one of the best opportunities, I
think, to do that, at least in a bipartisan way, is the
Platform Accountability and Consumer Transparency or the PACT
Act and that's legislation that I've co-sponsored with Senator
Schatz, which, in addition to stripping Section 230 protections
for content that a court determines to be illegal, the PACT Act
would also increase transparency and due process for users
around the content moderation process.
Importantly, in the context we're talking about today with
this hearing with a major Big Tech whistleblower, the PACT Act
would explore the viability of a Federal program for Big Tech
employees to blow the whistle on wrong-doing inside the
companies where they work.
In my view, we should encourage employees in the tech
sector like you to speak up about questionable practices of Big
Tech companies so we can, among other things, ensure that
Americans are fully aware of how social media platforms are
using artificial intelligence and opaque algorithms to keep
them hooked on the platform.
So let me, Ms. Haugen, just ask you. We've learned from the
information that you've provided that Facebook conducts what's
called ``engagement-based ranking,'' which you've described as
very dangerous.
Could you talk more about why engagement-based ranking is
dangerous and do you think Congress should seek to pass
legislation like the Filter Bubble Transparency Act that would
give users the ability to avoid engagement-based ranking all
together?
Ms. Haugen. Facebook is going to say you don't want to give
up engagement-based ranking. You're not going to like Facebook
as much if we're not picking out the content for you. That's
just not true. There are a lot of--Facebook likes to present
things as false choices, like you have to choose between having
lots of spam. Like let's imagine we ordered our feeds by time.
Like on I Message or on their other forms of social media that
are chronologically based. They're going to say you're going to
get spammed, like you're not going to enjoy your feed.
The reality is that those experiences have a lot of
permutations. There are ways that we can make those experiences
where computers don't regulate what we see. We together
socially regulate what we see, but they don't want us to have
that conversation because Facebook knows that when they pick
out the content, we focus on using computers, we spend more
time on their platform, they make more money.
The dangers of engagement-based ranking are that Facebook
knows that content that elicits an extreme reaction from you is
more likely to get a click, a comment, a reshare, and it's
interesting because those clicks and comments and reshares
aren't even necessarily for your benefit. It's because they
know that other people will produce more content if they get
the likes and comments and reshares.
They prioritize content in your feed so that you will give
little hits of dopamine to your friends so they will create
more content and they have run experiments on people, producer
side experiments where they have confirmed this.
Senator Thune. So you and part of the information you
provided the Wall Street Journal, it has been found that
Facebook altered its algorithm in an attempt to boost these
meaningful social interactions or MSI but rather than
strengthening bonds between family and friends on the platform,
the algorithm instead rewarded more outrage and sensationalism,
and I think Facebook would say that its algorithms are used to
connect individuals with other friends and family that are
largely positive.
Do you believe that Facebook's algorithms make its platform
a better place for most users and should consumers have the
option to use Facebook and Instagram without being manipulated
by algorithms designed to keep them engaged on that platform?
Ms. Haugen. I strongly believe =-like I've spent most of my
career working on systems like engagement-based ranking. Like
when I come to you and say these things, I'm basically damning
10 years of my own work, right.
Engagement-based ranking, Facebook says we can do it safely
because we have AI. You know, the artificial intelligence will
find the bad content that we know our engagement-based ranking
is promoting. They've written blog posts on how they know
engagement-based ranking is dangerous, but the AI will save us.
Facebook's own research says they cannot adequately
identify dangerous content and as a result, those dangerous
algorithms that they admit are picking up the extreme
sentiments, the division, they can't protect us from the harms
that they know exist in their own system and so I don't think
it's just a question of saying should people have the option of
choosing to not be manipulated by their algorithms.
I think if we had appropriate oversight or if we reformed
230 to make Facebook responsible for the consequences of their
intentional ranking decisions, I think they would get rid of
engagement-based ranking because it is causing teenagers to be
exposed to more anorexia content. It is pulling families apart
and in places like Ethiopia, it's literally fanning ethnic
violence.
I encourage reform of these platforms, not picking and
choosing individual ideas, instead making the platforms
themselves safer, less twitchy, less reactive, less viral,
because that's how we scalably solve these problems.
Senator Thune. Thank you.
Mr. Chair, I would simply say let's get to work. So we got
some things we can do here. Thanks.
Senator Blumenthal. I agree. Thank you.
Senator Schatz.
STATEMENT OF HON. BRIAN SCHATZ,
U.S. SENATOR FROM HAWAII
Senator Schatz. Thank you, Mr. Chairman, Ranking Member.
Thank you for your courage in coming forward. Was there a
particular moment when you came to the conclusion that reform
from the inside was impossible and that you decided to be a
whistleblower?
Ms. Haugen. It was a long series of moments where I became
aware that Facebook conflicts of interests between its own
profits and the common good public safety that Facebook
consistently chose to prioritize its profits.
I think the moment which I realized we needed to get help
from the outside, that the only way these problems would be
solved is by solving them together, not solving them alone, was
when Civic Integrity was dissolved following the 2020 election.
It really felt like a betrayal of the promises that
Facebook had made to people who had sacrificed a great deal to
keep the election safe by basically dissolving our community
and integrating into just other parts of the company.
Senator Schatz. And I know their response is that they sort
of distributed the duties.
Ms. Haugen. Yes.
Senator Schatz. That's an excuse, right?
Ms. Haugen. I cannot see into the hearts of other men and I
don't know what----
Senator Schatz. Let me say it this way. It won't work,
right?
Ms. Haugen. I can tell you that when I left the company, so
the people who I worked with were disproportionately maybe 75
percent of my pod of seven people, those are product managers,
program managers, most of them come from Civic Integrity. All
of us left the Inauthentic Behavior pod either for other parts
of the company or the company entirely over the same 6-week
period of time.
So 6 months after the reorganization, we had clearly lost
faith that those changes were coming.
Senator Schatz. You said in your opening statement that
``they know how to make Facebook and Instagram safer.'' So
thought experiment, you are now the chief executive officer and
chairman of the company. What changes would you immediately
institute?
Ms. Haugen. I would immediately establish a policy of how
to share information and research from inside the company with
appropriate oversight bodies, like Congress. I would give
proposed legislation to Congress, saying here's what an
effective oversight agency would look like. I would actively
engage with academics to make sure that people who are
confirming our Facebook's marketing message is true, have the
information they need to confirm these things, and I would
immediately implement the ``soft interventions'' that were
identified to protect the 2020 election. So that's things like
requiring someone to click on a link before resharing it
because other companies, like Twitter, have found that that
significantly reduces misinformation.
No one is censored by being forced to click on a link
before resharing it.
Senator Schatz. Thank you. I want to pivot back to
Instagram's targeting of kids.
We all know that they announced a pause but that reminds me
of what they announced when they were going to issue a digital
currency and they got beat up by the U.S. Senate Banking
Committee and they said never mind and now they're coming back
around hoping that nobody notices that they are going to try to
issue a currency.
Now let's set aside for the moment the business model which
appears to be gobble up everything, do everything, that's the
growth strategy.
Do you believe that they're actually going to discontinue
Instagram Kids or they're just waiting for the dust to settle?
Ms. Haugen. I would be sincerely surprised if they do not
continue working on Instagram Kids and I would be amazed if a
year from now we don't have this conversation again.
Senator Schatz. Why?
Ms. Haugen. Facebook understands that if they want to
continue to grow, they have to find new users, they have to
make sure that the next generation is just as engaged with
Instagram as the current one, and the way they'll do that is by
making sure that children establish habits before they have
good self-regulation.
Senator Schatz. By hooking kids?
Ms. Haugen. By hooking kids. I'd like to emphasize one of
the documents that we sent in on problematic use examined the
rates of problematic use by age and that peaked with 14-year-
olds. It's just like cigarettes. Teenagers don't have good
self-regulation. They say explicitly I feel bad when I use
Instagram and yet I can't stop. We need to protect the kids.
Senator Schatz. Just my final question. I have a long list
of misstatements, misdirections, and outright lies from the
company. I don't have the time to read them but you're as
intimate with all of these deceptions as I am. So I will just
jump to the end.
If you were a member of this panel, would you believe what
Facebook is saying?
Ms. Haugen. I would not believe. Facebook has not earned a
right to just have blind trust in them. Trust is--last week one
of the most beautiful things I heard on the committee was trust
is earned and Facebook has not earned our trust.
Senator Schatz. Thank you.
Senator Blumenthal. Thanks, Senator Schatz. Senator Moran,
and then we've been joined by the Chair, Senator Cantwell.
She'll be next.
We're going to break at about 11:30, if that's OK, because
we have a vote and then we'll reconvene.
Ms. Haugen. OK.
STATEMENT OF HON. JERRY MORAN,
U.S. SENATOR FROM KANSAS
Senator Moran. Mr. Chairman, thank you.
The conversation so far reminds me that you and I ought to
resolve our differences and introduce legislation. So as
Senator Thune said, let's go to work.
Senator Blumenthal. Our differences are very minor or they
seem very minor in the face of the revelations that we've now
seen. So I'm hoping we can move forward, Senator Moran.
Senator Moran. I share that view, Mr. Chairman. Thank you.
Thank you very much for your testimony. What examples do
you know--we've talked about particularly children, teenage
girls specifically, but what other examples do you know about
where Facebook or Instagram knew its decisions would be harmful
to its users but still proceeded with the plan and executed
that harmful behavior?
Ms. Haugen. Facebook's internal research is aware that
there are a variety of problems facing children on Instagram
that are--they know that severe harm is happening to children.
For example, in the case of bullying, Facebook knows that
Instagram dramatically changes the experience of high school.
So when we were in high school, when I was in high school, most
kids----
Senator Moran. You looked at me and changed your words.
[Laughter.]
Ms. Haugen. Sorry. When I was in high school, you know, or
most kids have positive home lives, like it doesn't matter how
bad it is at school, kids can go home and reset for 16 hours.
Kids who are bullied on Instagram, the bullying follows them
home. It follows them into their bedrooms. The last thing they
see before they go to bed at night is someone being cruel to
them or the first thing they see in the morning is someone
being cruel to them.
Kids are learning that their own friends, like people who
they care about them are cruel to them. Like think about how
that's going to impact their domestic relationships when they
become 20-somethings or 30-somethings to believe that people
who care about you are mean to you.
Facebook knows that parents today, because they didn't
experience these things, they never experienced this addictive
experience with a piece of technology, they give their children
bad advice. They say things like ``why don't you just stop
using it'' and so that Facebook's own research is aware that
children express feelings of loneliness and struggling with
these things because they can't even get support from their own
parents.
I don't understand how Facebook can know all these things
and not escalate it to someone like Congress for help and
support in navigating these problems.
Senator Moran. Let me ask the question in a broader way.
Besides teenagers and besides girls or besides youth, are there
other practices at Facebook or Instagram that are known to be
harmful but yet are pursued?
Ms. Haugen. Facebook is aware that choices it made in
establishing like ``meaningful social interactions,'' so
engagement-based ranking that didn't care if you bullied
someone or made hate speech in the comments that was
meaningful.
They know that that change directly changed publishers'
behavior, that companies like BuzzFeed wrote in and said the
content is most successful on our platform is some of the
content we're most ashamed of. You have a problem with your
ranking and they did nothing. They know that politicians are
being forced to take positions they know their own constituents
don't like or approve of because those are the ones that get
distributed on Facebook. That's a huge, huge negative impact.
Facebook also knows that they have admitted in public that
engagement-based ranking is dangerous without integrity and
security systems but then not rolled out those integrity and
security systems to most of the languages in the world and
that's what's causing things like ethnic violence in Ethiopia.
Senator Moran. Thank you for your answer. What is the
magnitude of Facebook's revenues or profits that come from the
sale of user data?
Ms. Haugen. Oh, I'm sorry, I've never worked on that. I'm
not aware.
Senator Moran. Thank you. What regulations or legal actions
by Congress or by administrative action do you think would have
the most consequence or be feared most by Facebook, Instagram,
or allied companies?
Ms. Haugen. I strongly encourage reforming Section 230 to
exempt decisions about algorithms, right. So modifying 230
around content, I think, it's very complicated because user-
generated content is something that companies have less control
over. They have one hundred percent control over their
algorithms and Facebook should not get a free pass on choices
it makes to prioritize growth and virility and reactiveness
over public safety. They shouldn't get a free pass on that
because they're paying for their profits right now with our
safety. So I strongly encourage reform of 230 in that way.
I also believe there needs to be a dedicated oversight body
because right now the only people in the world who are trained
to analyze these experiments to understand what's happening
inside of Facebook are people who, you know, grew up inside of
Facebook or Pinterest or another social media company, and
there needs to be a regulatory home where someone like me could
do a tour of duty after working at a place like this and have a
place to work on things like regulation to bring that
information out to the oversight boards that have the right to
do oversight.
Senator Moran. A regulatory agency within the Federal
Government?
Ms. Haugen. Yes.
Senator Moran. Thank you very much. Thank you, Chairman.
Senator Blumenthal. Senator Cantwell. Thank you, Senator
Moran.
STATEMENT OF HON. MARIA CANTWELL,
U.S. SENATOR FROM WASHINGTON
The Chairwoman. Thank you, Mr. Chairman. Thank you for
holding this hearing, and I think my colleagues have brought up
a lot of important issues and so I think I just want to
continue on that vein.
First of all, the Privacy Act that I introduced along with
several of my colleagues actually does have FTC oversight of
algorithm transparency in some instances. I hope you take a
look at that and tell us what other areas you think we should
add to that level of transparency.
But clearly that's the issue at hand here, I think, in your
coming forward. So thank you again for your willingness to do
that.
The documentation that you say now exists is the level of
transparency about what's going on that people haven't been
able to see and so your information that you say has gone up to
the highest levels at Facebook is that they purposely knew that
their algorithms were continuing to have misinformation and
hate information and that when presented with information about
this terminology, you know, downstream MSI, meaningful social
information, knowing that it was this choice, you could
continue this wrong-headed information, hate information about
the Rohingya or you could continue to get higher click-through
rates, and I know you said you don't know about profit, but I'm
pretty sure you know that on a page, if you click through that
next page, I'm pretty sure there's a lot more ad revenue than
if you didn't click through.
So you're saying the documents exist that at the highest
level at Facebook you had information discussing these two
choices and that people chose, even though they knew that it
was misinformation and hurtful and maybe even causing people
lives, they continued to choose profit?
Ms. Haugen. We have submitted documents to Congress
outlining Mark Zuckerberg was directly presented with a list of
``soft interventions.'' So hard intervention is like taking a
piece of content off Facebook, taking a user off Facebook. Soft
interventions are about making slightly different choices to
make the platform less viral, less twitchy.
Mark was presented with these options and chose to not
remove downstream MSI in April 2020 in even just isolated and
at-risk countries, that's countries at risk of violence, if it
had any impact on the overall MSI metric. So he chose----
The Chairwoman. Which, in translation, means less money,--
--
Ms. Haugen. Yes, he said----
The Chairwoman.--right? Was there another reason given why
they would do it, other than they thought it would really
affect their numbers?
Ms. Haugen. I don't know for certain. Like Jeff Horowitz,
the reporter at the Wall Street Journal, and I struggled with
this. We sat there and read these minutes and were like how is
this possible, like we've just read a hundred pages on how
downstream MSI expands hate speech, misinformation, violence-
inciting content, graphic violent content, why wouldn't you get
rid of this, and the best theory that we've come up with, and I
want to emphasize this is just our interpretation on it, is
people's bonuses are tied to MSI, right. Like people stay or
leave the company based on what they get paid and like if you
hurt MSI, a bunch of people weren't going to get their bonuses.
The Chairwoman. So you're saying that this practice even
still continues today? Like we're still in this environment?
I'm personally----
Ms. Haugen. Oh, yes.
The Chairwoman.--very frustrated by this because we
presented information to Facebook from one of my own
constituents in 2018 talking about this issue with Rohingya,
pleading with the company. We pleaded with the company and they
continued to not address this issue.
Now you're pointing out that these same algorithms are
being used and they know darn well in Ethiopia that it's
causing and inciting violence and again they are still today
choosing profit over taking this information down, is that
correct?
Ms. Haugen. When rioting began in the United States in the
summer of last year, they turned off downstream MSI only for
when they detected content was health content, which is
probably COVID, and civic content, but Facebook's own
algorithms are bad at finding this content. It's still in the
raw form for 80-90 percent of even that sensitive content.
In countries where they don't have integrity systems in the
local language and in the case of Ethiopia, there is a hundred
million people in Ethiopia and six languages. Facebook only
supports two of those languages for integrity systems. This
strategy of focusing on language-specific content-specific
systems AI to save us is doomed to fail.
The Chairwoman. I need to get to one of--first off fall,
I'm sending a letter to Facebook today. They better not delete
any information as it relates to the Rohingya or investigations
about how they proceeded on this particularly in light of your
information or the documents.
But aren't we also now talking about advertising fraud?
Aren't you selling something to advertisers that's not really
what they're getting? We know about this because of the
newspaper issues. We're trying to say that journalism that
basically has to meet a different standard, a public interest
standard that basically is out there basically proving every
day or they can be sued.
These guys are a social media platform that doesn't have to
live with that and then the consequences, they're telling their
advertisers that this was a--we see it. We see it. People are
coming back to the local journalism because they're like we
want to be again with the trusted brand. We don't want to be
in, you know, your website.
So I think you're finding for the SEC is an interesting
one, but I think that we also have to look at what are the
other issues here and one of them is did they defraud
advertisers in telling them this was the advertising content
that you were going to be advertising and when, in reality, it
was something different. It was based on a different model.
Ms. Haugen. We have multiple examples of question and
answers for the advertising staff, the sales staff where
advertisers say after the riots last summer were asked should
we come back to Facebook or after the insurrection, like should
we come back to Facebook, and Facebook said in their talking
points that they gave to advertisers we're doing everything in
our power to make this safer or we take down all the hate
speech when we find it.
The Chairwoman. That was not true.
Ms. Haugen. That was not true.
The Chairwoman. Thank you.
Ms. Haugen. They get three to 5 percent of hate speech.
The Chairwoman. Thank you. Thank you, Mr. Chairman.
Senator Blumenthal. Thanks, Senator Cantwell, and if you
want to make your letter available to other members of the
Committee, I'd be glad to join you myself----
The Chairwoman. Thank you, thank you.
Senator Blumenthal.--and thank you for suggesting it.
The Chairwoman. Thank you.
Senator Blumenthal. Senator Lee.
STATEMENT OF HON. MIKE LEE,
U.S. SENATOR FROM UTAH
Senator Lee. Thank you, Mr. Chairman, and thank you, Ms.
Haugen, for joining us this week. It's very helpful. We're
grateful that you're willing to make yourself available.
Last week we had another witness from Facebook, Ms. Davis.
She came and she testified before this committee and she
focused on, among other things, the extent to which Facebook
targets ads to children, including ads that are either sexually
suggestive or geared toward adult-themed products or themes in
general.
Now while I appreciated her willingness to be here, I
didn't get the clearest answers in response to some of those
questions and so I'm hoping that you can help shed some light
on some of those issues related to Facebook's advertising
processes here today.
As we get into this, I want to first read you a quote that
I got from Ms. Davis last week. Here's what she said during her
questioning. ``When we do ads to young people, there are only
three things that an advertiser can target around, age, gender,
location. We also prohibit certain ads to young people,
including weight loss ads. We don't allow tobacco ads at all.
We don't allow them to children. We don't allow them to
minors.''
Now since that exchange happened last week, there are a
number of individuals and groups, including a group called The
Technology Transparency Project or TTP, that have indicated
that that part of her testimony was inaccurate, that it was
false. TTP noted that TTP had conducted an experiment just last
month and their goal was to run a series of ads that would be
targeted to children ages 13 to 17, to users in the United
States.
Now I want to emphasize that TTP didn't end up running
these ads. They stopped them from being distributed to users,
but Facebook did in fact approve them, and as I understand it
Facebook approved them for an audience of up to 9.1 million
users, all of whom were teens.
So I brought a few of these to show you today. This is the
first one I wanted to showcase. This first one has a colorful
graphic encouraging kids to ``throw a Skittles party like no
other,'' which, as the graphic indicates and as the slang
jargon also independently suggests, this involves kids getting
together randomly to abuse prescription drugs.
The second graphic displays an anitip, that is, a tip
specifically designed to encourage and promote anorexia, and
it's on there. Now the language, the anitip itself,
independently promotes that. The ad also promotes it insofar as
it was suggesting these are images you ought to look at when
you need motivation to be more anorexic, I guess you could say.
Now the third one invites children to find their partner
online and to make a love connection. You look lonely, find
your partner now to make a love connection. Now, look, it could
be an entirely different kettle of fish if this were targeted
to an adult audience. It is not. It's targeted to 13- to 17-
year-olds.
Now obviously I don't support and TTP does not support
these messages, particularly when targeted to impressionable
children, and again just to be clear, TTP did not end up
pushing the ads out after receiving Facebook's approval, but it
did in fact receive Facebook's approval.
So I think this says something, one could argue that it
proves that Facebook is allowing and perhaps facilitating the
targeting of harmful adult-themed ads to our Nation's children.
So could you please explain to me, Ms. Haugen, how these
ads with the target audience of 13-to-17-year-old children, how
would they possibly be approved by Facebook? Is AI involved in
that?
Ms. Haugen. I did not work directly on the ad approval
system. What was resonant for me about your testimony is
Facebook has a deep focus on scale. So scale is can we do
things very cheaply for a huge number of people which is part
of why they rely on AI so much.
It is very possible that none of those ads were seen by a
human and the reality is, as we've seen from repeated documents
within my disclosures, is that Facebook's AI systems only catch
a very tiny minority of offending content and best case
scenario in the case of something like hate speech at most they
will ever get 10 to 20 percent.
In the case of children, that means drug paraphernalia ads
like that, it's likely if they rely on computers and not humans
they will also likely never get more than 10 to 20 percent of
those ads.
Senator Lee. Mr. Chairman, I've got one minor follow-up
question. It should be easy to answer.
Senator Blumenthal. Go ahead.
Senator Lee. So while Facebook may claim that it only
targets ads based on age, gender, and location, even though
these things seem to counteract that, but let's set that aside
for a minute, and that they're not basing ads based on specific
interest categories, does Facebook still collect interest
category data on teenagers, even if they aren't at that moment
targeting ads at teens based on those interest categories?
Ms. Haugen. I think it's very important to differentiate
between what targeting are advertisers allowed to specify and
what targeting Facebook may learn for an ad.
Let's imagine you have some texts on an ad. It would likely
extract out features that it thought was relevant for that ad.
For example, in the case of something about partying, it would
learn partying is a concept.
I'm very suspicious that personalized ads are still not
being delivered to teenagers on Instagram because the
algorithms learn correlations. They learn interactions where
your party ad may still go to kids interested in partying
because Facebook almost certainly has a ranking model in the
background that says this person wants more party-related
content.
Senator Lee. Interesting. Thank you. That's very helpful
and what that suggests to me is that while they're saying
they're not targeting teens with those ads, the algorithm might
do some of that work for them which might explain why they
collect the data even while claiming that they're not targeting
those ads in that way.
Ms. Haugen. I can't speak to whether or not that's the
intention, but the reality is, it's very, very, very difficult
to understand these algorithms today and over and over and over
again we saw these biases the algorithms unintentionally learn
and so, yes, it's very hard to disentangle out these factors as
long as we have engagement-based ranking.
Senator Lee. Thank you, Ms. Haugen.
Senator Blumenthal. Thank you very much, Senator Lee.
Senator Markey.
STATEMENT OF HON. EDWARD MARKEY,
U.S. SENATOR FROM MASSACHUSETTS
Senator Markey. Thank you, Mr. Chairman, very much.
Thank you, Ms. Haugen. You are a 21st Century American hero
warning our country of the danger for young people, for our
democracy, and our Nation owes you just a huge debt of
gratitude for the courage you're showing here today. So thank
you.
Ms. Haugen, do you agree that Facebook actively seeks to
attract children and teens on to its platforms?
Ms. Haugen. Facebook actively markets to children or
markets to children under the age of 18 to get on Instagram and
definitely targets children as young as eight to be on
Messenger Kids.
Senator Markey. An internal Facebook document from 2020
that you revealed reads, ``Why do we care about tweens? They
are a valuable but untapped audience.'' So Facebook only cares
about children to the extent that they are of monetary value.
Last week Facebook's Global Head of Safety, Antigone Davis,
told me that Facebook does not allow targeting of certain
harmful content to teens. Ms. Davis stated, ``We don't allow
weight loss ads to be shown to people under the age of 18.''
Yet a recent study found that Facebook permitted targeting of
teens as young as 13 with ads that showed a young woman's thin
waist promoting websites that glorify anorexia.
Ms. Haugen, based on your time at Facebook, do you think
Facebook is telling the truth?
Ms. Haugen. I think Facebook has focused on scale over
safety and it is likely that they are using artificial
intelligence to try to identify harmful ads without allowing
the public oversight to see what is the actual effectiveness of
those safety systems.
Senator Markey. You unearthed Facebook's research about its
harm to teens. Did you raise this issue with your supervisors?
Ms. Haugen. I did not work directly on anything involving
teen mental health. This research is freely available to anyone
in the company.
Senator Markey. Ms. Davis testified last week, ``We don't
allow tobacco ads at all. We don't allow them to children
either. We don't allow alcohol ads to minors.'' However,
researchers also found that Facebook does allow targeting of
teens with ads on vaping.
Ms. Haugen, based on your time at Facebook, do you think
Facebook is telling the truth?
Ms. Haugen. I do not. I have context on that issue. I
assume that if they are using artificial intelligence to catch
those vape ads, unquestionably ads are making its way through.
Senator Markey. OK. So from my perspective listening to you
and your incredibly courageous revelations, time and time again
Facebook says one thing and does another. Time and time again
Facebook fails to abide by the commitments that they have made.
Time and time again Facebook lies about what they are doing.
Yesterday Facebook had a platform outage but for years it
has had a principles outage. Its only real principle is profit.
Facebook's platforms are not safe for young people. As you
said, Facebook is like Big Tobacco, enticing young kids with
that first cigarette, that first social media account designed
to hook kids as users for life.
Ms. Haugen, your whistleblowing shows that Facebook uses
harmful features that quantify popularity, push manipulative
influencer marketing, amplify harmful content to teens, and
last week in this committee Facebook wouldn't even commit to
not using these features on 10-year-olds. Facebook is built on
computer codes of misconduct.
Senator Blumenthal and I have introduced the Kids Internet
Design and Safety Act, the KIDS Act. You have asked us to act
as a committee and Facebook has scores of lobbyists in this
city right now coming in right after this hearing to tell us we
can't act and they have been successful for a decade in
blocking this committee from acting.
So let me ask you a question. The Kids Internet Design and
Safety Act or the KIDS Act, here's what the legislation does.
It includes outright bans on children's app features that: (1)
quantify popularity with likes and follower accounts, (2)
promotes influencer marketing, and (3) that amplifies of toxic
posts and that it would prohibit Facebook from using its
algorithms to promote toxic posts. Should we pass that
legislation?
Ms. Haugen. I strongly encourage reforms that push us
toward human-scale social media and not computer-driven social
media. Those amplification harms are caused by computers
choosing what's important to us, not our friends and family,
and I encourage any system that children are exposed to to not
use amplification systems.
Senator Markey. So you agree that Congress has to enact
these special protections for children and teens to stop social
media companies from manipulating young users and threatening
their well-being, to stop using its algorithm to harm kids? You
agree with that?
Ms. Haugen. I do believe Congress must act to protect
children.
Senator Markey. And children and teens also need privacy,
online bill of rights. I'm the author of the Children's Online
Privacy Protection Act of 1998, but it's only for kids under 13
because the industry stopped me from making it age 16 in 1998
because it was already their business model, but we need to
update that law for the 21st Century.
Tell me if this should pass: (1) create an online eraser
button so that young users can tell websites to delete the data
they have collected about them, (2) give young teens under the
age of 16 and their parents control of their information, and
(3) ban targeted ads to children.
Ms. Haugen. I support all those actions.
Senator Markey. Thank you. And finally I've also introduced
the Algorithmic Justice and Online Platform Transparency Act
which would: (1) open the hood on Facebook and Big Tech's
algorithms so we know how Facebook is using our data to decide
what content we see and (2) ban discriminatory algorithms that
harm vulnerable populations online, like showing employment and
housing ads to white people but not to black people in our
country.
Should Congress pass that bill?
Ms. Haugen. Algorithmic bias issues are a major issue for
our democracy. During my time at Pinterest, I became very aware
of the challenges of--like I mentioned before, it's difficult
for us to understand how these algorithms actually act and
perform.
Facebook is aware of complaints today by people like
African Americans saying that reals doesn't give African
Americans the same distribution as white people and until we
have transparency and our ability to confirm ourselves that
Facebook's marketing messages are true, we will not have a
system that is compatible with democracy.
Senator Markey. So I thank Senator Lee, I really do, and
your line of questioning. I wrote Facebook asking them to
explain that discrepancy because Facebook, I think, is lying
about targeting 13-year- to-year-olds.
So here's my message for Mark Zuckerberg. Your time of
invading our privacy, promoting toxic content, and preying on
children and teens is over. Congress will be taking action. You
can work with us or not work with us, but we will not allow
your company to harm our children and our families and our
democracy any longer.
Thank you, Ms. Haugen. We will act.
Senator Blumenthal. Thanks, Senator Markey.
We're going to turn to Senator Blackburn and then we will
take a break. I know that there's some interest in another
round of questions. Maybe we'll turn to Senator Lujan for his
questions.
Senator Blackburn. We have Senators Cruz and Scott.
Senator Blumenthal. And we have others. So we'll come back
after the----
Senator Lujan. Mr. Chairman, I have to go sit in the chair
starting at noon.
Senator Blumenthal. Why don't we turn to--you have
questions?
Senator Blackburn. I do. I have one question. This relates
to what Mr. Markey was asking.
Does Facebook ever employ child psychologists or mental
health professionals to deal with these children online issues
that we're discussing?
Ms. Haugen. Facebook has many researchers with PhDs. I
assume some of them are--I know that some have psychology
degrees. I'm not sure if they are child specialists. Facebook
also works with external agencies that are specialists at
children's rights online.
Senator Blackburn. Thank you.
Senator Blumenthal. Senator Lujan, and then at the
conclusion of Senator Lujan's questions, we'll take a break.
We'll come back at noon.
STATEMENT OF HON. BEN RAY LUJAN,
U.S. SENATOR FROM NEW MEXICO
Senator Lujan. Thank you, Mr. Chairman, and I appreciate
the indulgence of the committee.
Ms. Haugen, last week the committee heard directly from Ms.
Davis, the Global Head of Safety for Facebook. During the
hearing the company contested their own internal research as if
it does not exist.
Yes or no. Does Facebook have internal research indicating
that Instagram harms teens, particularly harming perceptions of
body image, which disproportionately affects young women?
Ms. Haugen. Yes, Facebook has extensive research on the
impacts of its products on teenagers, including young women.
Senator Lujan. Thank you for confirming these reports.
Last week I requested Facebook make the basis of this
research, the Dataset Minus, any personally identifiable
information available to this committee.
Do you believe it is important for transparency and safety
that Facebook release the basis of this internal research, the
core dataset, to allow for independent analysis?
Ms. Haugen. I believe it is vitally important for our
democracy that we establish mechanisms where Facebook's
internal research must be disclosed to the public on a regular
basis and that we need to have privacy-sensitive datasets that
allow independent researchers to confirm whether or not
Facebook's marketing messages are actually true.
Senator Lujan. Beyond this particular research, should
Facebook make its internal primary research, not just secondary
slide decs of cherry-picked data, but the underlying data
public by default? Can this be done in a way that respects user
privacy?
Ms. Haugen. I believe in collaboration with academics and
other researchers that we can develop privacy-conscious ways of
exposing radically more data that is available today. It is
important for our ability to understand how algorithms work,
how Facebook shapes the information we get to see, that we have
these datasets be publicly available for scrutiny.
Senator Lujan. Is Facebook capable of making the right
decision here on its own or is regulation needed to create real
transparency at Facebook?
Ms. Haugen. Until incentives change at Facebook, we should
not expect Facebook to change. We need action from Congress.
Senator Lujan. Last week I asked Ms. Davis about shadow
profiles for children on the site and she answered that no data
is ever collected on children under 13 because they are not
allowed to make accounts. This tactfully ignores the issue.
Facebook knows children uses their platform. However,
instead of seeing this as a problem to be solved, Facebook
views this as a business opportunity.
Yes or no. Does Facebook conduct research on children under
13, examining the business opportunities of connecting these
young children to Facebook's products?
Ms. Haugen. I want to emphasize how vital it is that
Facebook should have to publish the mechanisms by which it
tries to detect these children because they are on the platform
in far greater numbers than anyone is aware.
I do believe and I am aware that Facebook is doing research
on children under the age of 13 and those studies are included
in my disclosure.
Senator Lujan. You have shared your concerns about how
senior management at Facebook has continuously prioritized
revenue over potential user harm and safety, and I have a few
questions on Facebook's decisionmaking.
Last week I asked Ms. Davis, ``Has Facebook ever found a
change to its platform would potentially inflict harm on users
but Facebook moved forward because the change would also grow
users or increase revenue?'' Ms. Davis said in response, ``It's
not been my experience at all at Facebook. That's just not how
we would approach it.''
Yes or no. Has Facebook ever found a feature on its
platform harmed its users but the feature moved forward because
it would also grow users or increase revenue?
Ms. Haugen. Facebook likes to paint that these issues are
really complicated. There are lots of simple issues. For
example, requiring someone to click through on a link before
you reshare it, that's not a large imposition but it does
decrease growth a tiny little amount because in some countries
reshares make up 35 percent of all the content that people see.
Facebook prioritized that content on the system, the
reshares, over the impacts to misinformation, hate speech, or
violence incitement.
Senator Lujan. Did these decisions ever come from Mark
Zuckerberg directly or from other senior management at
Facebook?
Ms. Haugen. We have a few choice documents that contain
notes from briefings with Mark Zuckerberg where he chose
metrics defined by Facebook, like meaningful social
interactions, over changes that would have significantly
decreased misinformation, hate speech, and other inciting
content.
Senator Lujan. And this is the reference you shared earlier
to Ms. Cantwell, April 2020.
Ms. Haugen. Yes, the soft interventions.
Senator Lujan. Facebook appeared to be able to count on the
silence of its workforce for a long time, even as it knowingly
continued practices and policies that continued to cause and
amplify harm. Facebook content moderators have called out ``a
culture of fear and secrecy within the company that prevented
them from speaking out.''
Is there a culture of fear at Facebook around
whistleblowing and external accountability?
Ms. Haugen. Facebook has a culture that emphasizes that
insularity is the path forward, that if information is shared
with the public, it will just be misunderstood, and I believe
that relationship has to change. The only way that we will
solve these problems is by solving them together and we will
have much better, more democratic solutions if we do it
collaboratively than in isolation.
Senator Lujan. And my final question, is there a senior
level executive at Facebook, like an inspector general, who's
responsible for ensuring complaints from Facebook employees are
taken seriously and that employees' legal, ethical, and moral
concerns receive consideration with the real possibility of
instigating change to company policies?
Ms. Haugen. I am not aware of that role, but the company is
large and it may exist.
Senator Lujan. I appreciate that. It's my understanding
that there's a gentleman by the name of Roy Austin who is the
Vice President of Civil Rights who's described himself as an
inspector general, but he does not have the authority to make
these internal conflicts public.
The Oversight Board was created by Facebook to review
moderation policies related to public content specifically. It
was not created to allow employees to raise concerns. So again
another area of interest I believe that we have to act on.
I thank you for coming forward today.
Ms. Haugen. My pleasure. Happy to serve.
Senator Blumenthal. The Committee is in recess.
[Recess.]
Senator Blumenthal. Welcome back, Ms. Haugen. Thank you for
your patience.
We're going to reconvene and we'll go to Senator
Hickenlooper.
STATEMENT OF HON. JOHN HICKENLOOPER,
U.S. SENATOR FROM COLORADO
Senator Hickenlooper. Thank you, Mr. Chair, and thank you,
Ms. Haugen, for your direct answers and for being willing to
come out and, you know, provide such clarity on so many of
these issues.
Obviously Facebook can manipulate its algorithms to attract
users and I guess my question would be do you feel in your
humble opinion that, you know, simply maximizing profits no
matter the societal impact that that is justified and I think
the question then would be--that's the short question, which I
think I know the answer.
What impact to Facebook's bottom line would it have if the
algorithm was changed to promote safety and instead changed to
save the lives of young women rather than putting them at risk?
Ms. Haugen. Facebook today makes approximately $40 billion
a year in profit. A lot of the changes that I'm talking about
are not going to make Facebook an unprofitable company. It just
won't be a ludicrously profitable company like it is today.
Engagement-based ranking, which causes those amplification
problems that leads young women from, you know, innocuous
topics like healthy recipes to anorexia content, if it were
removed, people would consume less content on Facebook, but
Facebook would still be profitable and so I encourage oversight
and public scrutiny into how these algorithms work and the
consequences of them.
Senator Hickenlooper. And I appreciate that. I'm a former
small business owner and started a brew pub back in 1988 and we
worked very hard to look--again, we weren't doing
investigations but we were very sensitive to whether someone
had had too much to drink, whether we had a frequent customer
who was frequently putting himself at risk and others.
Obviously I think the Facebook business model puts--well,
poses risk to youth and to teens. You compared it to cigarette
companies which I thought was rightfully so.
I guess the question is, is this level of risk appropriate
or is there a level of risk that would be appropriate?
Ms. Haugen. I think there is an opportunity to reframe some
of these oversight actions. So when we think of them as these
tradeoffs of like it's either profitability or safety, I think
that's a false choice, and in reality, the thing I'm asking for
is a move from short-termism, which is what Facebook is run
under today, right, is being led by metrics and not led by
people, and that with appropriate oversight and some of these
constraints, it's possible that Facebook could actually be a
much more profitable company five or 10 years down the road
because it wasn't as toxic, not as many people quit it, but
that's one of those counter-factuals that we can't actually
test.
So regulation might actually make Facebook more profitable
over the long term.
Senator Hickenlooper. Right. That's often the case. I think
the same could be said for automobiles and go down the list----
Ms. Haugen. Definitely.
Senator Hickenlooper.--of all those things. There's so much
pushback in the beginning.
I also thought that the question of how do we assess the
impact to their bottom line. We had a representative of
Facebook in here recently who talked about that eight out of 10
Facebook users feel their life is better and that their job is
to get to 10 out of 10. Maybe this is the 20 percent that
they're missing. I don't know how large the demographic is of
people that are caught back up into this circuitous, you know,
sense of really taking them down into the wrong direction, how
many people that is. Do you have any idea?
Ms. Haugen. That quote last week was really shocking to me
because I don't know if you're aware of this but in the case of
cigarettes, only about 10 percent of people who smoke ever get
lung cancer, right. So the idea that, you know, 20 percent of
your users could be facing serious mental health issues and
that's not a problem is shocking.
I also want to emphasize for people that eating disorders
are serious, right. There are going to be women walking around
this planet in 60 years with brittle bones because of choices
that Facebook made around emphasizing profit today or there are
going to be women in 20 years who want to have babies who can't
because they're infertile as a result of eating disorders
today. They're serious, and I think there's an opportunity here
for having public oversight and public involvement, especially
in matters that impact children.
Senator Hickenlooper. Well, thank you for being so direct
on this and for stepping forward. I yield back the floor, Mr.
Chair.
Senator Blumenthal. Thanks, Senator Hickenlooper.
Senator Cruz.
STATEMENT OF HON. TED CRUZ,
U.S. SENATOR FROM TEXAS
Senator Cruz. Thank you, Mr. Chairman.
Ms. Haugen, welcome. Thank you for your testimony.
When it concerns Facebook, there are a number of concerns
that this committee and Congress has been focused on. Two of
the biggest have been Facebook's intentional targeting of kids
with content that is harmful to the children and then, second,
a discreet issue is the pattern of Facebook and social media
engaging in political censorship.
I want to start with the first issue, targeting kids. As
you're aware and as, indeed, the documents that you provided
indicated, Facebook, according to the public reporting on it,
Facebook's internal reports found that Instagram makes ``body
image issues worse for one in three teen girls,'' and
additionally it showed that ``13 percent of British users and 6
percent of American users traced their desire to kill
themselves to Instagram.''
Is that a fair and accurate characterization of what
Facebook's research concluded?
Ms. Haugen. I only know what I read in the documents that
were included in my disclosure. That is an accurate description
of the ones that I have read. Because Facebook has not come
forward with the total corpus of their known research, I don't
know what their other things say, but, yes, there is documents
that say those things.
Senator Cruz. So at testimony last week in the Senate with
the witness from Facebook who claimed that that information was
not accurate and needed to be in context, of course, she wasn't
willing to provide the context, the alleged mysterious context,
do you know of any context that would make those data anything
other than horrifying and deeply disturbing?
Ms. Haugen. Engagement-based ranking and these processes of
amplification, they impact all users of Facebook. The
algorithms are very smart in the sense that they latch on to
things that people want to continue to engage with and,
unfortunately, in the case of teen girls and things like self-
harm, they develop these feedback cycles where children are
using Instagram as to self-soothe but then are exposed to more
and more content that makes them hate themselves.
This is a thing where we can't say 80 percent of kids are
OK. We need to say how do we save all the kids.
Senator Cruz. The Wall Street Journal reported that Mark
Zuckerberg was personally aware of this research. Do you have
any information one way or the other as to Mr. Zuckerberg's
awareness of the research?
Ms. Haugen. One of the documents included in the
disclosures details something called Project Daisy, which is an
initiative to remove likes off of Instagram. The internal
research showed that removing likes off Instagram is not
effective as long as you leave comments on those posts and yet
the research directly presented to Mark Zuckerberg said we
should still pursue this as a feature to launch even though
it's not effective because the government, journalists, and
academics want us to do this, like it would get us positive
points with the public.
That kind of duplicity is why we need to have more
transparency and why if we want to have a system that is
coherent with democracy we must have public oversight from
Congress.
Senator Cruz. Do you know if Facebook, any of the research
it conducted, attempted to quantify how many teenage girls may
have taken their lives because of Facebook's products?
Ms. Haugen. I'm not aware of that research.
Senator Cruz. Do you know if Facebook made any changes when
they got back that 13 percent of British users and 6 percent of
American users traced their desire to kill themselves to
Instagram? Do you know if they made any changes in response to
that research to try to correct or mitigate that?
Ms. Haugen. I found it very surprising that when Antigone
Davis was confronted with this research last week she couldn't
enumerate a five-point plan, a 10-point plan of the actions
that they took.
I also find it shocking that once Facebook had this
research it didn't disclose it to the public because this is
the kind of thing that should have oversight from Congress.
Senator Cruz. So when you were at Facebook were there
discussions about how to respond to this research?
Ms. Haugen. I did not work directly on issues concerning
children. These are just documents that were freely available
in the company. So I'm not aware of that.
Senator Cruz. OK. Do you have thoughts as to what kind of
changes Facebook could make to reduce or eliminate these harms?
Ms. Haugen. You mentioned earlier concerns around free
speech. A lot of the things that I advocate for are around
changing the mechanisms of amplification, not around picking
winners and losers in the marketplace of ideas. So problems----
Senator Cruz. Explain what that means.
Ms. Haugen. Oh, sure. So like I mentioned before, you know,
like how on Twitter if you have to click through on a link
before you reshare it, small actions like that friction don't
require picking good ideas and bad ideas. They just make the
platform less twitchy, less reactive, and Facebook's internal
research says that each one of those small actions dramatically
reduces misinformation, hate speech, and violence-inciting
content on the platform.
Senator Cruz. So we're running out of time, but on the
second major topic of concern of Facebook, which is censorship,
based on what you've seen, are you concerned about political
censorship at Facebook and in Big Tech?
Ms. Haugen. I believe you cannot have a system that has as
big an impact on society as Facebook does today with as little
transparency as it does. I'm a strong proponent of
chronological ranking, ordering by time with a little bit of
spam demotion because I think we don't want computers deciding
what we focus on. We should have software that is human-scaled
where humans have conversations together, not computers
facilitating who we get to hear from.
Senator Cruz. So how could we get more transparency? What
would produce that?
Ms. Haugen. I strongly encourage the development of some
kind of regulatory body that could work with academics, work
with researchers, work with other government agencies to
synthesize requests for data that are privacy-conscious.
This is an area that I'm really passionate about because
right now no one can force Facebook to disclose data and
Facebook has been stonewalling us or, even worse, they gave
inaccurate data to researchers as the scandal recently showed.
Senator Cruz. What data should they turn over? My time has
expired.
Ms. Haugen. For example, even data as simple as what
integrity systems exist today and how well do they perform.
Like there are lots and lots of people who Facebook is
conveying around the world that Facebook's safety systems apply
to their language and those people aren't aware that they're
using a raw original dangerous version of Facebook. Just basic
actions like transparency would make a huge difference.
Senator Cruz. Thank you.
Senator Blumenthal. Thanks, Senator Cruz.
Senator Lummis.
STATEMENT OF HON. CYNTHIA LUMMIS,
U.S. SENATOR FROM WYOMING
Senator Lummis. Thank you, Mr. Chairman, and thank you for
your testimony.
If you were in my seat today instead of your seat, what
documents or unanswered questions, would you seek from
Facebook, especially as it relates to children, but even
generally speaking?
Ms. Haugen. I think any research regarding what Facebook
does problematic use, i.e., the addictiveness of the product,
is of vital importance and anything around what Facebook knows
about parents' lack of knowledge about the platform.
I only know about the documents that I have seen, right. I
do not work on teens or child safety myself, but in the
documents that I read, Facebook articulates the idea that
parents today are not aware of how dangerous Instagram is and
because they themselves do not live through these experiences,
they can't coach their kids on basic safety things and so at a
minimum Facebook should have to disclose what it knows in that
context.
Senator Lummis. OK. So we're trying to protect individuals'
data that they're gathering, have data privacy, but have
transparency in the manner in which the data is used. Can we
bridge that gap?
Ms. Haugen. Imagine--I think reasonable people can have a
conversation on how many people need to see a piece of content
before it's not really private. Like if a 100,000 people see
something, is it private? If 25,000 people see it, is it
private?
Just disclosing the most popular content on the platform,
including statistics around what factors went into the
promotion of that content, would cause radically more
transparency than we have today on how Facebook chooses what we
get to focus on, how they shape our reality.
Senator Lummis. OK. If our focus is protecting the First
Amendment and our rights to free speech while very carefully
regulating data privacy, there are a number of things that are
being discussed in Congress, everything from antitrust laws to
calling Facebook a utility to the idea that you just raised of
a regulatory board of some sort that has authority through
understanding of the algorithms and how they're used and other
mechanisms that create what we see, the face of Facebook, so to
speak.
Tell me a little more about how you envision that board
working. What is the--in your mind, based on your understanding
of the company and the ill consequences, what is the best
approach to bridging the gap between keeping speech free and
protecting individual privacy with regard to data?
Ms. Haugen. So I think those issues, they're independent
issues. So we can talk about free speech first which is having
more transparency--like Facebook has solutions today that are
not content-based and I am a strong advocate for non-content-
based solutions because those solutions will also then protect
the most vulnerable people in the world.
In a place like Ethiopia where they speak six languages, if
you have something that focuses on good ideas and bad ideas,
those systems don't work in diverse places.
So investing in non-content-based ways to slow the platform
down not only protects our freedom of speech, it protects
people's lives.
The second question is around privacy and it's a question
of how can we have oversight and have privacy. There is lots
and lots of research on how to extract datasets so you're not
showing people's names. You might not even be showing the
content of their posts. You might be showing data that is about
the content of their posts but not the post itself.
There are many ways to structure these datasets that are
privacy-conscious and the fact that Facebook has walled off the
ability to see even basic things about how the platform
performs or in the case of their past academic research
releasing inaccurate data or not being clear about how they
pulled that data is just part of a pattern of behavior of
Facebook hiding behind walls and operating in the shadows and
they have far too much power in our society to be allowed to
continue to operate that way.
Senator Lummis. Well, I had heard you make the analogy
earlier to the tobacco industry and I think that that's an
appropriate analogy. I really believe we're searching for the
best way to address the problem, and I'm not sure that it is
the heavy hands, like breaking up companies or calling them a
utility, which is why your approach of integrating people who
understand the math and the uses of the math with protecting
privacy is intriguing to me.
So the more information that you can provide to us about
how that might work to actually address the problem, I think
would be helpful. So in my case, this is an invitation to you
to provide to my office or the committee information about how
we can get at the root of the problem that you've identified
and can document and save people's privacy.
So I extend that invitation to you and I thank you for your
testimony.
Ms. Haugen. Thank you.
Senator Lummis. Mr. Chairman, I yield back.
Senator Blumenthal. Thanks, Senator Lummis.
Senator Sullivan.
STATEMENT OF HON. DAN SULLIVAN,
U.S. SENATOR FROM ALASKA
Senator Sullivan. Thank you, Mr. Chairman, and I want to
thank our witness here. Been a good hearing, a lot of
information has been learned, particularly on the issue of how
this is impacting our kids.
I think we're going to look back 20 years from now and all
of us are going to be like what in the hell were we thinking
when we recognize the damage that's done to a generation of
kids. Do you agree with that, Ms. Haugen?
Ms. Haugen. When Facebook has made statements in the past
about how much benefit Instagram is providing to kids' mental
health, like kids are connecting who were once alone, why I'm
so surprised about what is if Instagram is such a positive
force, have we seen a golden age of teenage mental health in
the last 10 years? No. We've seen----
Senator Sullivan. We've seen the opposite, right?
Ms. Haugen. We've seen escalating rates of suicide and
depression amongst teenagers.
Senator Sullivan. Do you think those rates are at least in
part driven by the social media phenomena?
Ms. Haugen. There is a broad swath of research that
supports the idea that usage of social media amplifies the risk
for these mental health harms.
Senator Sullivan. Right now this hearing is helping to
illuminate it. We are seeing----
Ms. Haugen. And Facebook's own research shows that.
Senator Sullivan. Right. Say that again. That's important.
Ms. Haugen. And Facebook's own research shows that, right,
that kids are saying I am unhappy when I use Instagram and I
can't stop, that if I leave I'm afraid I'll be ostracized.
Senator Sullivan. Right.
Ms. Haugen. And that's so sad.
Senator Sullivan. So they know that.
Ms. Haugen. That's what the research shows.
Senator Sullivan. So what do you think drives them to--I
had this discussion with the witness last week and I said,
well, you know, I think they called it their time-out or stop.
I said but isn't that incompatible with your business model
because your business model is more time online, more eyeballs
online? Isn't that the fundamental elements of their business
model?
Ms. Haugen. Facebook has had both an interesting
opportunity and a hard challenge from being a closed system. So
they have had the opportunity to hide their problems and like
often people do when they can hide their problems, they get in
over their heads, and I think Facebook needs an opportunity to
have Congress step in and say guess what, you don't have to
struggle by yourself anymore, you don't have to hide these
things from us, you don't have to keep pretending they're not
problems. You can declare moral bankruptcy and we can figure
out how to fix these things together because we solve problems
together. We don't solve them alone.
Senator Sullivan. And by moral bankruptcy, one of the
things that I appreciate the phrase that the Chairman and you
have been using, is one of those elements which is they know
this is a problem, they know it's actually impacting negatively
the mental health of the most precious assets we have in
America, our youth, our kids, I have three daughters. They know
that that is happening and yet the moral bankruptcy from your
perspective is the continuation of this simply because that's
how they make money.
Ms. Haugen. I phrased it slightly differently. We have a
financial bankruptcy because we value people's lives more than
we value money. The people get in over their heads and they
need a process where they admit they did something wrong, but
we have a mechanism where we forgive them and we have a way for
them to move forward.
Facebook is stuck in a feedback loop that they cannot get
out of. They have been hiding this information because they
feel trapped, right, like they would have come forward if they
had solutions to these things. They need to admit they did
something wrong and they need help to solve these problems and
that's what moral bankruptcy is.
Senator Sullivan. Let me ask--I'm going to switch gears
here and this is--what's your current position right now in
terms of disinformation and counterespionage?
Ms. Haugen. My last role at Facebook was in
Counterespionage.
Senator Sullivan. I'm sorry. Your last role. OK.
Ms. Haugen. Yes.
Senator Sullivan. So one of the things--this is a very
different topic and I only got a minute or so left, but right
now is Facebook--I know Facebook is not allowed in countries
like China, but do they provide platforms for authoritarian or
terrorist-based leaders, like the Ayatollahs in Iran, that's
the largest state-sponsored terrorism in the world, or the
Taliban or Xi Jinping or certain, in my view, our biggest rival
for this century, a Communist Party dictator who's trying to
export his authoritarian model around the world, do they
provide a platform for those kind of leaders who, in my view,
clearly don't hold America's interests in mind? Does Facebook
provide that platform?
Ms. Haugen. During my time working with the Threat
Intelligence Org, so I was a product manager supporting the
Counterespionage Team, my team directly worked on tracking
Chinese participation on the platform, surveilling, say, weaker
populations in places around the world. You could actually find
the Chinese based on them doing these kinds of things.
Senator Sullivan. So Facebook--I'm sorry.
Ms. Haugen. We also saw active participation of, say, the
Iran Government doing espionage on other state actors. So this
is definitely a thing that is happening and I believe
Facebook's consistent understaffing of the Counter-espionage
Information Operations and Counterterrorism Teams is a national
security issue and I'm speaking to other parts of Congress
about that.
Senator Sullivan. So you are saying in essence that the
platform, whether Facebook knows it or not, is being utilized
by some of our adversaries in a way that helps push and promote
their interests at the expense of America's?
Ms. Haugen. Yes, Facebook's very aware that this is
happening on the platform, and I believe the fact that Congress
doesn't get a report of exactly how many people are working on
these things internally is unacceptable because you have a
right to keep the American people safe.
Senator Sullivan. Great. Thank you very much.
Senator Blumenthal. Thanks, Senator Sullivan. You may have
just opened an area for another hearing.
Ms. Haugen. Sorry. Yes, I've strong national security
concerns about how Facebook operates today.
Senator Sullivan. Well, Mr. Chairman, maybe we should,
right. I mean, it's a real issue.
Senator Blumenthal. I'm not being at all facetious. Thank
you for your questions on this topic and I know you have a busy
schedule, but we may want to discuss this issue with you with
members of our committee at least informally and if you'd be
willing to come back for another hearing. That certainly is
within the realm of possibility. I haven't consulted with the
Ranking Member or the Chairwoman, but thank you for your
honesty and your candor on that topic.
Senator Scott.
STATEMENT OF HON. RICK SCOTT,
U.S. SENATOR FROM FLORIDA
Senator Scott. Thank you, Chairman.
First off, thanks for coming forward and thanks for coming
forward in the manner that you wanted to have positive change.
So that's not always what happens.
Earlier this year I sent a letter to Facebook and other
social media platforms asking them to detail the harmful
impacts that affects our mental health their platforms have on
children and teens.
So your report revealed that Facebook has been clearly
fully aware of this for awhile and the harmful impacts,
especially on young women. So I think we all agree that's
completely unacceptable and we've got to figure out how we
protect the people that are vulnerable in this country from the
harmful impacts of Facebook and other social media platforms.
So, first off, do you think there should be greater
consideration for age when it comes to using any social media?
Ms. Haugen. I strongly encourage raising age limits to 16
or 18 years old based on looking at the data around problematic
use or addiction on the platform and children's self-regulation
issues.
Senator Scott. So I think you addressed this a little bit,
but why do you think Facebook didn't address this publicly when
they figured out internally that they were having an adverse
impact on young people, especially young women? Why didn't they
come forward and say we've got a problem, we've got to figure
this out?
Ms. Haugen. I have a huge amount of empathy for Facebook.
These are really, really hard questions and part of why I say--
I think they feel a little trapped and isolated is the problems
that are driving negative social comparison on Instagram,
Facebook's own research says Instagram is actually distinctly
worse than, say, TikTok or Snapchat or Reddit because
Instagram--TikTok is about doing fun things with your friends.
Snapchat is about faces and augmented reality. Reddit is
vaguely about ideas, but Instagram is about bodies and about
comparing lifestyles and so I think there are real questions
where like Instagram would have to come in and think hard about
their product or about like what is their product about, and I
don't feel those answers are immediately obvious.
That's why I believe we need to solve problems together and
not alone because collaborating with the public will give us
better solutions.
Senator Scott. So do you think Facebook was trying to
mitigate the problem?
Ms. Haugen. I think within the set of incentives that they
were working within, they did the best they could.
Unfortunately, those incentives are not sustainable and they
are not acceptable in our society.
Senator Scott. Do you think Facebook and other social media
platforms ought to be required to report any harmful effects
they have on young people?
Ms. Haugen. One of the things I found very interesting
after the report in the Wall Street Journal on teen mental
health was that a former executive at the company said,
``Facebook needs to be able to have private research,'' and the
part that I was offended by this was Facebook has had some of
this research on the negative effects of Instagram on teenagers
for years. I strongly support the idea that Facebook should
have a year, maybe 18 months to have private research, but
given that they are the only people in the world who can do
this kind of research, the public never gets to do it, they
shouldn't be allowed to keep secrets when people's lives are on
the line.
Senator Scott. So----
Ms. Haugen. Because to be clear, if they make $40 billion a
year, they have the resources to solve these problems. They're
choosing not to solve them.
Senator Scott.--does that surprise you, they wouldn't put
more effort into this?
Ms. Haugen. No.
Senator Scott. They knew it was going to catch up with them
eventually, right?
Ms. Haugen. Yes.
Senator Scott. Why wouldn't they----
Ms. Haugen. Like I mentioned earlier to Senator
Hickenlooper, coming in and having oversight might actually
make Facebook a more profitable company five or 10 years from
now because toxicity, Facebook's own research shows they have
something called an integrity holdout. These are people who
don't get protections from integrity systems to see what
happens to them and those people who deal with the more toxic
painful version of Facebook use Facebook less and so one could
reason a kinder, friendlier, more collaborative Facebook might
actually have more users 5 years from now. So it's in
everyone's interests.
Senator Scott. Do you think--I've got a bill and there are
a lot of bills that I think we've all talked about, but mine's
called the Data Act. It's going to require expressed consent
from users for large platforms to use algorithms on somebody.
Do you agree with that? I mean, shouldn't we consent before
they get to take everything about us and go sell it and how
they send things to us?
Ms. Haugen. For selling personal data, that is an issue I
believe people should have substantially more control over.
Most people are not well informed on what the costs, the
personal costs of having their data sold are and so I worry
about pushing that choice back on individual consumers.
In terms of should people consent to working with
algorithms, I worry that if Facebook is allowed to give users
the choice of do you want an engagement-based newsfeed or do
you want a chronological newsfeed, like ordered by time, maybe
a little spam demotion, that people will choose the more
addictive option, that engagement-based ranking, even if it is
leading their daughters to eating disorders.
Senator Scott. All right. Thank you.
Senator Blumenthal. Thanks, Senator Scott.
I think we have concluded the first round, unless we're
missing someone who is online and not hearing anyone, let's go
to the second round.
Thank you again for your patience. I know you have a hard
stop, I think, at 1:30. So we'll be respectful of that
limitation and I'll begin by asking a few questions.
First, let me say Senator Klobuchar very aptly raised with
you the principle obstacle to our achieving legislative reform
in the past, which is the tons of money spent on lobbyists and
other kinds of influence peddling, to use a pejorative word
that is so evident here in the U.S. Congress. Some of it is
dark money, some of it is very overt, but I guess the point I
would like to make to you personally is that your being here
really sends a profound message to our Nation that one person
can really make a difference, one person standing up, speaking
out, can overcome a lot of those obstacles for us, and you have
crystallized in a way our consciousness here, you have been a
catalyst, I think, for change in a way that we haven't seen,
and I've been working on these issues for 10-15 years, and you
have raised awareness in a way that I think is very unique.
So thank you not only for your risk-taking and your courage
and strength in standing up but also for the effect that it has
had, and I also want to make another point. You can tell me
whether I'm correct or not.
I think there are other whistleblowers out there. I think
there are other truth-tellers in the tech world who want to
come forward, and I think you are leading by example. I think
you are showing them that there is a path to make this industry
more responsible and more caring about kids and about the
nature of our public discourse generally and about the strength
of our democracy, and I think you have given them a boost,
those whistleblowers out there, in potentially coming forward.
I think that is tremendously important and I think also, again
you can tell me if I am wrong, there are a lot of people in
Facebook who are cheering for you because there are public
reports and I know of some of my friends in this world who tell
me that there are people working for Facebook who wish they had
the opportunity and the courage to come forward as you have
done because they feel a lot of reservations about the way that
Facebook has used the platform, used algorithms, used content
and pushed it on kids in this way.
So those are sort of hypotheses that I hope you can confirm
and I also would like to ask you because a lot of parents are
watching right now. So you've advised us on what you think we
should do: the reforms, some of them that you think we should
adopt, stronger oversight authorized by Congress, better
disclosure because right now Facebook essentially is a black
box----
Ms. Haugen. YES.
Senator Blumenthal.--for most of America. Facebook is a
black box that's designed by Mark Zuckerberg, Incorporated,
Mark Zuckerberg and his immediate coterie, and the buck stops
with him, and reform of Section 230, so there's some legal
responsibility, so people have a day in court, some kind of
recourse, legally, when they're harmed by Facebook because
right now it has this broad immunity. Most of America has no
idea.
Essentially you can't sue Facebook. You have no recourse.
Most of America doesn't know about Section 230 and if you
pushed a lot of Members of Congress, they wouldn't know either.
Ms. Haugen. It's actually slightly worse than that.
Facebook made a statement in a legal proceeding recently where
they said they had the right to mislead the court because they
had immunity, right, that 230----
Senator Blumenthal. --Exactly----
Ms. Haugen.--gave them immunity, so why should they have to
tell the truth about what they're showing?
Senator Blumenthal. Which is kind of----
Ms. Haugen. Shocking.
Senator Blumenthal.--concerning. Well, it is shocking to a
lawyer, which some of us are. It is also utter disregard and
contempt for the Rule of Law and for the very legal structure
that gives them that kind of protection. So it's kind of a new
low in corporate conduct at least in court.
So you have provided us with some of the reforms that you
think are important and I think that the oversight goes a long
way because it in turn would make public a lot of what is going
on in this black box.
But for now, since a lot of teens and tweens will be going
home tonight, as you have said, to endure the bullying, the
eating disorders, the invitations to feel insecure about
themselves, heightened anxiety, they have to live with the real
world as it exists right now, and they will be haunted for
their lifetimes by these experiences.
What would you tell parents right now? What would you
advise them about what they can do because they need more tools
and some of the proposals that have been mentioned here would
give parents more tools to protect their children? Right now a
lot of parents tell me they feel powerless. They need more
information. They are way behind their kids in their adeptness
online and they feel that they need to be empowered in some way
to protect their kids in the real world right now in real time.
So I offer you that open-ended opportunity to talk to us a
little bit about your thoughts.
Ms. Haugen. Very rarely do you have one of these
generational shifts where the generation that leads, like
parents who guide their children, have such a different set of
experiences that they do not have the context to support their
children in a safe way.
There is an active need for schools or maybe the National
Institutes of Health to make established information where if
parents want to learn how they can support their kids, it
should be easy for them to know what is constructive and not
constructive because Facebook's own research says kids today
feel like they are struggling alone with all these issues
because their parents can't guide them.
One of the things I am saddest is when I look on Twitter is
when people blame the parents for these problems with Facebook.
They say just take your kid's phone away and the reality is the
issues are a lot more complicated than that.
And so we need to support parents because right now if
Facebook won't protect the kids, we at least need to help the
parents to protect the kids.
Senator Blumenthal. Parents are anguished----
Ms. Haugen. They are.
Senator Blumenthal.--about this issue. Parents are hardly
uncaring. They need the tools. They need to be empowered and I
think that the major encouragement for reforms is going to come
from those parents and you have pointed out, I think in general
but I'd like you to just confirm for me, this research and the
documents containing that research is not only findings and
conclusions, it's also recommendations for changes.
What I hear you saying is that again and again and again
these recommendations were just rejected or disregarded,
correct?
Ms. Haugen. There is a pattern of behavior that I saw at
Facebook of Facebook choosing to prioritize its profits over
people and any time that Facebook faced even tiny hits to
growth, like .1 percent of sessions, 1 percent of views, that
it chose its profits over safety.
Senator Blumenthal. And you mentioned, I think, bonuses
tied to downstream MSIs.
Ms. Haugen. To core MSI, yes.
Senator Blumenthal. Could you explain what you meant?
Ms. Haugen. So MSI is Meaningful Social Interaction.
Facebook's internal governance is very much based around
metrics. So Facebook is incredibly flat to the point where they
have the largest open floor plan office in the world. It's a
quarter of a mile long and one room, right, they believe in
flat.
Instead of having internal governance, they have metrics
that people try to move. In a world like that, it doesn't
matter that we now have multiple years of data saying MSI may
be encouraging bad content, might be making spaces where people
are scared, where they are shown information that puts them at
risk. It's so hard to dislodge a ruler like that that--a
yardstick, that you end up in a situation where because no one
is taking leadership, like no one is intentionally designing
these systems, it's just many, many people running in parallel,
all moving the metric, that these problems get amplified and
amplified and amplified and no one steps in to bring the
solutions.
Senator Blumenthal. And I just want to finish and then I
think we've been joined by Senator Young, and then we'll go to
Senator Blackburn and Senator Klobuchar.
You know, I spent a number of years as an Attorney General
helping to lead litigation against Big Tobacco and I came to
hear from a lot of smokers how grateful they were, ironically
and unexpectedly, that someone was fighting Big Tobacco because
they felt they had been victimized as children. They started
smoking when they were seven, eight, 12-years-old because Big
Tobacco was hooking them and as we developed the research very
methodically and purposefully addicting them at that early age
when they believed that they would make themselves more
popular, that they would be cool and hip if they began smoking
and then nicotine hooked them.
Now physiologically nicotine has addictive properties. What
is it about Facebook's tactics of hooking young people that
makes it similar to what Big Tobacco has done?
Ms. Haugen. Facebook's own research about Instagram
contains quotes from kids saying I feel bad when I use
Instagram but I also feel like I can't stop, right. I know that
the more time I spend on it, the worse I feel, but like I just
can't--like they want the next click, they want the next like,
the dopamine, the little hits all the time, and I feel a lot of
pain for those kids, right. Like they say they fear being
ostracized if they step away from the platform. So imagine
you're in the situation, in this relationship where every time
you open the app, it makes you feel worse, but you also fear
isolation if you don't.
I think there's a huge opportunity here to make social
media that makes kids feel good, not feel bad, and that we have
an obligation to our youth to make sure that they're safe
online.
Senator Blumenthal. Thank you.
Senator Young.
STATEMENT OF HON. TODD YOUNG,
U.S. SENATOR FROM INDIANA
Senator Young. Ms. Haugen, thank you for your compelling
testimony.
In that testimony you discuss how Facebook generates self-
harm and self-hate, especially among vulnerable groups, like
teenage girls. I happen to be a father of four kids, three
daughters, two of whom are teenagers, and as you just alluded
to, most adults, myself included, have never been a teenager
during the age of Facebook, Instagram, and these other social
media platforms, and therefore I think it can be really hard
for many of us to fully appreciate the impact that certain
posts may have, including, I would add, on the teens' mental
health.
Can you discuss the short- and long-term consequences of
body image issues on these platforms, please?
Ms. Haugen. The patterns that children establish in their
teenage years live with them for the rest of their lives. The
way they conceptualize who they are, how they conceptualize how
they interact with other people are patterns and habits they
will take with them as they become adults, as they themselves
raise children.
I'm very scared about the upcoming generation because when
you and I interact in person and I say something mean to you
and I see you wince or I see you cry, that makes me less likely
to do it the next time, right. That's a feedback cycle.
Online kids don't get those cues and they learn to be
incredibly cruel to each other and they normalize it, and I'm
scared of what will their lives look like where they grow up
with the idea that it's OK to be treated badly by people who
allegedly care about them. That's a scary future.
Senator Young. Very scary future, and I see some evidence
of that as to so many parents on a regular basis.
Are there other specific issues of significant consequence
that the general public may not be fully aware of that are
impacting vulnerable groups that you'd just like to elevate
during this testimony?
Ms. Haugen. One of the things that's hard for people who
don't look at the data of social networks every day, it can be
hard to conceptualize the distribution patterns of harms or
just of usage. There are these things called power logs. That
means that a small number of users are extremely intensely
engaged on any given topic and most people are just lightly
engaged.
When you look at things like misinformation, Facebook knows
that the people who are exposed to the most misinformation are
people who are recently widowed, divorced, moved to a new city,
are isolated in some other way.
When I worked on Civic Misinformation, we discussed the
idea of the misinformation burden, like the idea that when
people are exposed to ideas that are not true over and over
again, it arose their ability to connect with the community at
large because they no longer adhere to facts that are consensus
reality.
The fact that Facebook knows that its most vulnerable
users, people who are recently widowed, like that they're
isolated, that the systems that are meant to keep them safe,
like demoting this information, stop working when people look
at 2,000 posts a day, right.
It breaks my heart the idea that these rabbit holes would
suck people down and then make it hard to connect with others.
Senator Young. So, Ms. Haugen, I desperately want to, which
is the American impulse, I want to solve this problem and----
Ms. Haugen. Me, too.
Senator Young.--I very much believe that Congress not only
has a role but has a responsibility to figure this out. I don't
pretend to have all the answers.
I would value your opinion, though, as to whether you
believe that breaking up Facebook would solve any of the
problems that you've discussed today. Do you think it would?
Ms. Haugen. So as an algorithmic specialist, so someone who
designs algorithmic experiences, I'm actually against the
breaking up of Facebook because even looking inside of just
Facebook itself, not even Facebook and Instagram, you see the
problems of engagement-based ranking repeat themselves.
So the problems here are about the design of algorithms, of
AI, and the idea that AI is not intelligent, and if you break
up Instagram and Facebook from each other, it's likely--so I
used to work on Pinterest and the thing that we faced from a
business model perspective was that advertisers didn't want to
learn multiple advertising platforms, that they wanted to
learn--they got one platform for Instagram and Facebook and
whatever and learning a second one for Pinterest, Pinterest
made radically fewer dollars per user.
What I'm scared of is right now Facebook is the Internet
for lots of the world. If you go to Africa, the Internet is
Facebook. If you split Facebook and Instagram apart, it's
likely that most advertising dollars will go to Instagram and
Facebook will continue to be this Frankenstein that is
altering--like that is endangering lives around the world, only
now there won't be money to fund it.
So I think oversight and finding collaborative solutions
with Congress is going to be key because these systems are
going to continue to exist and be dangerous even if broken up.
Senator Young. Thank you.
Senator Blumenthal. Thanks, Senator Young.
Senator Blackburn.
Senator Blackburn. Thank you, Mr. Chairman.
I have a text that was just put out by Facebook's
spokesperson, Mr. Stone. It says, ``Just pointing out the fact
that Frances Haugen did not work on child safety or Instagram
or research these issues and has no direct knowledge of the
topic from her work at Facebook.''
So I will simply say this to Mr. Stone. If Facebook wants
to discuss their targeting of children, if they want to discuss
their practices, privacy invasion, or violations of the
Children Online Privacy Act, I am extending to you an
invitation to step forward, be sworn in, and testify before
this committee. We would be pleased to hear from you and
welcome your testimony.
One quick question for you. What's the biggest threat to
Facebook's existence? Is it greed? Is it regulators? Is it
becoming extinct or obsolete for teenage users? What is the
biggest threat to their existence?
Ms. Haugen. I think the fact that Facebook is driven so
much by metrics and that these lead to a very heavy emphasis on
short-termism. Every little individual decision may seem like
it helps with growth, but if it makes it a more and more toxic
platform that people don't actually enjoy, like when they
passed meaningful social interactions back in 2018, Facebook's
own research said that users said it made it less meaningful,
right.
I think this aggregated set of short-term decisions
endangers Facebook's future, but sometimes we need to pull it
away from business as usual, help it write new rules if want it
to be successful in the future.
Senator Blackburn. So they can't see the forest for the
trees.
Ms. Haugen. Yes, yes.
Senator Blackburn. Thank you. And I know Senator Klobuchar
is waiting. So I'll yield my time back, and I thank you.
Senator Blumenthal. Thanks, Senator Blackburn.
Senator Klobuchar. Thanks very much, and thank you to both
of you for your leadership and all three of us are on the
Judiciary Committee, so we're also working on a host of other
issues, including the App Store issues which is unrelated to
Facebook actually, including issues relating to dominant
platforms when they promote their own content or engage in
exclusionary conduct, which I know is not our topic today.
I see the thumb's up from you, Ms. Haugen, which I
appreciate. I think this idea of establishing some rules of the
road for these tech platforms goes beyond the kid protection
that we so dearly need to do, and I just want to make sure you
agree with me on that.
Ms. Haugen. Yes, totally. I was shocked when I saw the New
York Times story a couple weeks ago about Facebook using its
own platform to promote positive news about itself. I was like
wow, a new shape to our reality. I wasn't aware it was that
much.
Senator Klobuchar. Right. And that's a lot of the work that
we're doing over there.
So I want to get to get something Senator Young was talking
about, misinformation, and Senator Lujan and I have put
together an exception actually to the 230 immunity when it
comes to vaccine misinformation in the middle of a public
health crisis.
Last week YouTube announced it was swiftly banning all
anti-vaccine misinformation and I've long called on Facebook to
take similar steps. They've taken some steps, but do you think
they can remove this content and do they put sufficient
resources?
We know the effect of this. We know that over half the
people that haven't gotten the vaccine just because of
something that they've seen on social media. I know the guy I
walked into a cafe and said his mother-in-law wouldn't get a
vaccine because she thought a microchip would be planted in her
arm, which is false. I'm just saying that now for the record
here----
Ms. Haugen. Yes.
Senator Klobuchar.--in case it gets put on social media.
Could you talk about are there enough resources to stop
this from happening?
Ms. Haugen. I do not believe Facebook as currently
structured has the capability to stop vaccine misinformation
because they overly rely on artificial intelligence systems
that they themselves say will likely never get more than 10 to
20 percent of content.
Senator Klobuchar. There you go. And yet it's a company
that what, over a trillion dollars, one of the world's biggest
companies that we've ever known and that's what really bothers
me.
Senator Lujan and I also have pointed out the issue with
content moderators. Does Facebook have enough content
moderators for content in Spanish and other languages besides
English?
Ms. Haugen. One of the things that is disclosed--we have
documentation that shows how much operational investment there
was by different languages and it showed a consistent pattern
of under-investment in languages that are not English.
I am deeply concerned about Facebook's ability to operate
in a safe way in languages beyond maybe the top 20 in the
world.
Senator Klobuchar. OK. Thank you. Go back to eating
disorders. Today you've said that you have documents indicating
Facebook is doing studies on kids under 13, even though
technically no kids under 13 are permitted on the platform.
The potential for eating disorder content to be shown to
these children raises serious concerns. Senator Blumenthal has
been working on this. I've long been focused on this eating
disorder issue, given the mortality rates.
Are you aware of studies Facebook has conducted about
whether kids under 13 on the platform are nudged toward content
related to eating disorders or unhealthy diet practices? CNN
also did investigation on this front.
Ms. Haugen. I have not seen specific studies regarding
eating disorders in under the age of 13, but I have seen
research that indicates that they are aware that teenagers
coach tweens who are on the platform to not reveal too much, to
not post too often, and that they've categorized that as a
``myth,'' that you can't be authentic on the platform, and that
the marketing team should try to advertise to teenagers to stop
coaching tweens that way. So I believe we shared that document
with Congress already.
Senator Klobuchar. Exactly. Well, thank you, and we'll be
looking more--speaking of the research issue, Facebook has
tried to downplay the internal research that was done, saying
it was unreliable.
It seems to me that they're trying to mislead us there. The
research was extensive, surveying hundreds of thousands of
people traveling around the world to interview users.
In your view, are the internal researchers at Facebook who
examined how users are affected by the platform, is their work
thorough? Are they experienced? Is it fair for Facebook to
throw them under the bus?
Ms. Haugen. Facebook has one of the top-ranked research
programs in the tech industry, like they've invested more in it
than I believe any other social media platform.
Some of the biggest heroes inside the company are the
researchers because they are boldly asking real questions and
being willing to say awkward truths. The fact that Facebook is
throwing them under the bus, I think, is unacceptable and I
just want the researchers to know that I stand with them and
that I see them.
Senator Klobuchar. Or maybe we should say, as the name of
one book, the ugly truth.
Ms. Haugen. Yes.
Senator Klobuchar. What about Facebook blocking researchers
at NYU from accessing the platform, does that concern you?
These are outside researchers.
Ms. Haugen. I am deeply concerned. So for context for those
who are not familiar with this research, there are researchers
at NYU who, because Facebook does not publish enough data on
political advertisements or how they are distributed, these are
advertisements that influence our democracy and how it
operates, they created a plug-in that allowed people to opt in,
to volunteer to help collect this data collectively, and
Facebook lashed out at them and even banned some of their
individual accounts.
The fact that Facebook is so scared of even basic
transparency that it goes out of its way to block researchers
who are asking awkward questions shows you the need for
congressional oversight and why we need to do Federal research
and Federal regulations on this.
Senator Klobuchar. Very good. Thank you. Thank you for your
work.
Senator Blumenthal. Thanks, Senator Klobuchar.
Senator Markey.
Senator Markey. Thank you. Thank you, Mr. Chairman. Thank
you for your incredible leadership on this issue.
As early as 2012, Facebook has wanted to allow children
under the age of 12 to use its platform. At that time in 2012 I
wrote a letter to Facebook asking questions about what data it
planned to collect and whether the company intended to serve
targeted ads at children. Now here we are 9 years later
debating the very same issues.
Today, Ms. Haugen, you've made it abundantly clear why
Facebook wants to bring more children on to the platform. It's
to hook them early, just like cigarettes, so that they become
lifelong users, so Facebook's profits increase.
Yet we should also ask why in the last 9 years has the
company not launched Facebook for Kids or Instagram for Kids?
After all, from the testimony here today, Facebook appears to
act without regard to any moral code or any conscience or
instead puts profit above people, profit above all else.
The reason why Facebook hasn't officially permitted kids 12
and under to use its platform is because the Child Online
Privacy Protection Act of 1998 that I'm the author of exists,
because there is a privacy law on the books which I authored
that gives the Federal Trade Commission regulatory power to
stop websites and social media companies from invading the
privacy of our children 12 and under.
That's why we need to expand the Child Online Privacy
Protection Act. That's why we need to pass the KIDS Act that
Senator Blumenthal and I have introduced, and why we need an
algorithmic justice act to pass because the absence of
regulation leads to harming teens, stoking division, damaging
our democracy. That's what you've told us today.
So, Ms. Haugen, I want you to come back to the protections
that you are calling on us to enact. This isn't complicated.
We're going to be told online all day with these paid Facebook
people, oh, Congress can't act. They're not experts. It's too
complicated for Congress. Just get out of the way. You're not
experts.
Well, this isn't complicated. Facebook and its Big Tech
lobbyists are blocking my bills to protect kids because it
would cost them money. That's how complicated it is.
So let's start with the KIDS Act that Senator Blumenthal
and I that would ban influencer marketing to kids. Today's
popular influencers peddle products while they flaunt their
lavish lifestyles to young users.
Can you explain how allowing influencer marketing to teens
and children makes Facebook more money?
Ms. Haugen. The business model that provides a great deal
of the content on Instagram is one where people produce content
for free. They put it on Instagram free. No one's charged for
it. But many of those content creators have sponsorships from
brands or from other affiliate programs.
Facebook needs those content creators to continue to make
content so that we will view content and in the process view
more ads.
Facebook provides tools to support influencers who do
influencer marketing because it gives them the supply of
content that allows them to keep people on the platform viewing
more ads, making more money for them.
Senator Markey. Yes. So I am actually the author of the
1990 Children's Television Act. What does that do? Well, it
says to all the television networks in America stop preying
upon children, stop using all of your power in order to try to
get young children in our country hooked on the products that
are going to be sold.
We had to pass a law to ban television stations from doing
this. That's why I knew that after my law passed in 1996 to
break up the monopolies of the telecommunications industry and
allow in the Googles and the Facebooks and all the other
companies, you name it, that we would need a child privacy
protection there because everyone would just move over to that
new venue.
It was pretty obvious and, of course, the industry said no
way we're going to have privacy laws for adults and they
blocked me from putting that on the books in 1996, but at least
for children I got up to age 12. That's all I could get out of
the industry.
But we also know that as time has moved on they've become
even more sophisticated so that the KIDS Act is necessary to
stop children and teen apps from being features, such as likes
and follower counts that quantify popularity.
Ms. Haugen, can you explain how allowing these features
that create an online popularity contest makes Facebook more
money?
Ms. Haugen. Just to make sure, so I am only familiar with
issues regarding teens from the research I have read of
Facebook's. So I want to put that caveat on there.
The research I have seen with regard to quantifiable
popularity is that as long as comments are allowed, so this is
not a quantitative thing, this is comments, as long as comments
are still on posts on Instagram, just taking likes off
Instagram doesn't fix the social comparison problem, that, you
know, teenage girls are smart. They see that Sally is prettier
than them, her pictures are really good, she gets tons of
comments, they don't ever make comments, right, and so I do
think we need larger interventions than just removing
quantitative measures.
Facebook has a product that is very attractive. The reason
why they have the study of problematic use is because it is
kind of addictive and those kinds of things, like having lots
of little feedback loops, keeps kids engaged and like I
mentioned earlier, part of why Facebook switched over to
meaningful social interactions was it found that if you got
more likes, more comments, more reshares, you produced more
content.
So having those systems of little rewards makes people
produce more content, which means we view more content and we
view more ads which makes them more money.
Senator Markey. OK. And the KIDS Act that Senator
Blumenthal and I are advocating for also prohibits
amplification of dangerous and violent content to children and
teens.
Can you explain how algorithms pushing that dangerous
content makes Facebook more money?
Ms. Haugen. I don't think Facebook ever set out to
intentionally promote divisive extreme polarizing content. I do
think, though, that they are aware of the side effects of the
choices they have made around amplification and they know that
algorithmic-based rankings, so engagement-based ranking, keeps
you on their sites longer. You have longer sessions. You show
up more often and that makes them more money.
Senator Markey. So do you believe we have to ban all
features that quantify popularity as a starting point in
legislation?
Ms. Haugen. As I covered before, the internal research I
have seen is that removing things like likes alone, if you
don't remove things like comments, doesn't have a huge impact
on social comparisons. So I do believe we need to have a more
integrated solution for these issues.
Senator Markey. OK. Should we ban targeted advertisements
to children?
Ms. Haugen. I strongly encourage banning targeted
advertisements to children and we need to have oversight in
terms of I think the algorithms will likely still learn the
interests of kids and match ads to those kids. Even if the
advertiser can't articulate, they'll want to target on those
interests.
Senator Markey. Right. How much money does Facebook make
from targeting children?
Ms. Haugen. Targeting children? I don't know what fraction
of the revenue comes from children.
Senator Markey. OK. So ultimately children are not
commodities.
Ms. Haugen. No.
Senator Markey. They've always been given historically
special protections. That's what the Children's Television Act
of 1990 is all about. They've always been given this special
safety zone so that children can grow up without being preyed
upon by marketers.
When I was a boy and the salesman would knock on the front
door, my mother would just say tell him I'm not home. That man
is not getting into our living room. Well, I would say to my
mother but you are home. Not to him, she would say.
Well, we need to give parents the ability just to say no
one's home for you and your company and your attempts to prey
upon children, to get into our living room. That's our moment
in history and we have to make sure that we respond to the
challenge.
Thank you, Mr. Chairman.
Senator Blumenthal. Thank you, Senator Markey, and my
thanks to Senator Markey for his leadership over many years on
protecting children. As you have heard, he was a champion in
the House of Representatives before coming here well before I
was in the U.S. Senate but around the time I was elected
Attorney General. I've been very pleased and honored to work
with him on legislation now going forward and I join him in
thanking you.
I have just a few concluding questions and I seem to be the
last one left standing here. So the good news is I don't think
we'll have others.
As you may know, you do know, my office created an
Instagram user identified as a 13-year-old girl. She followed a
few easily identifiable accounts on weight loss, dieting,
eating disorders, and she was deluged literally within a day of
content pushed to her by algorithms that in effect promoted
self-injury and eating disorders. Are you surprised by that
fact?
Ms. Haugen. I'm not surprised by that fact. Facebook has
internal research where they have done even more gentle
versions of that experiment, where they have started from
things like interest in healthy recipes, so not even extreme
dieting, and because of the nature of engagement-based ranking
and amplification of interests, that imaginary user was pushed
or that real account was pushed toward extreme dieting and pro-
anorexia content very rapidly.
Senator Blumenthal. And that's the algorithm.
Ms. Haugen. That's the algorithm.
Senator Blumenthal. That algorithm could be changed.
Ms. Haugen. The algorithm definitely could be changed. I
have firsthand experience from having worked at Pinterest.
Pinterest used to be an application that was heavily based just
on you following certain people's pings and those are put into
your feed and over time it grew to be much, much more heavily
based on recommendations that the algorithm would figure out
what are you interested in.
You can have wonderful experiences that are based on human
interactions. So these are human-scale technologies, not
computers choosing what we focus on.
Senator Blumenthal. So the average parent listening here
worried about their daughter or son being deluged with these
kinds of content would want that kind of algorithm changed, I
would think, and would welcome the oversight that you're
recommending.
Ms. Haugen. I believe parents deserve more options and more
choices, and today they don't know even what they could be
asking for.
Senator Blumenthal. I just received by text literally about
15 minutes ago a message from someone in Connecticut and I'm
going to read it to you. It is from a dad. ``I'm in tears right
now watching your interaction with Frances Haugen. My 15-year-
old daughter loved her body at 14, was on Instagram constantly
and maybe posting too much. Suddenly she started hating her
body, her body dysmorphia, now anorexia, and was in deep, deep
trouble before we found treatment. I fear she'll never be the
same. I'm broken-hearted.''
I think people tend to lose sight of the real world impact
here----
Ms. Haugen. Yes.
Senator Blumenthal.--and I think that is the reason that
you are here. I would just like to invite you, if you have any
words to those other employees at Big Tech, the workers who may
be troubled by the misconduct or unethical conduct that they
see, what you would tell them?
Ms. Haugen. We live in a pattern that we have seen
throughout time with regards to technologies. Humans are very
crafty people. We find interesting solutions. We often get out
over our skies, right. We develop things that are of a larger
scale than we really know how to handle, and what we have done
in the past is when we see this happen, we take a step back and
we find institutions and we find frameworks for doing these
things in a safe way.
We live in a moment where whistleblowers are very important
because these technological systems are walled off. They are
very complicated. They're things that you need to be a
specialist to really understand the consequences of and the
fact that we've been having the exact same kinds of false
choice discussions about what to do about Facebook, you know,
is it a privacy or oversight, is it about censorship or safety,
but the fact that we are being asked these false choices is
just an illustration of what happens when the real solutions
are hidden inside of companies.
We need more tech employees to come forward through
legitimate channels, like the FCC or Congress, to make sure
that the public has the information they need in order to have
technologies be human-centric, not computer-centric.
Senator Blumenthal. Thank you.
On that note, we'll conclude.
Ms. Haugen. Thank you.
Senator Blumenthal. Thank you for an extraordinary
testimony. I think that anybody watching would be impressed and
much better informed and you have done America a real public
service. I thank you.
Ms. Haugen. Thank you.
Senator Blumenthal. The record will remain open for two
weeks. Any Senators who want to submit questions for the record
should do so by October 19.
This hearing is adjourned.
Ms. Haugen. Thank you.
[Whereupon, at 1:22 p.m., the hearing was adjourned.]
A P P E N D I X
Response to Written Questions Submitted by Hon. Roy Blunt to
Frances Haugen
Question 1. In your testimony, you focused in particular on the
effects of Facebook's products on children, including in terms of
addiction and mental health. As a Senator who has heard from countless
Missouri parents on this very issue, and as a parent and grandparent
myself, this is a huge concern of mine as well. Children's screen time,
including on social media, has spiked in recent years--especially
during the pandemic--and we don't have an adequate picture of the
corresponding long- and short-term harms on children's health and
development. I think better research in this area is needed, both to
inform public policy and to help parents make educated decisions.
That's why, earlier this year, along with bipartisan colleagues from
the Senate and House, I co-sponsored the introduction of the CAMRA Act,
which would authorize the National Institutes of Health to conduct
research into the effects of technology and media, including social
media, on children in their cognitive, physical, and socio-emotional
development.
Ms. Haugen, as both a data scientist and Facebook alumni, do you
agree that the CAMRA Act would be a step in the right direction in
better understanding the effects of social media on children and to
help private companies design safer products? What advice would you
give NIH in conducting this study, to yield accurate, informative, and
useful research?
Answer. Thank you Senator Blunt, for your excellent questions and
your care and consideration of the issues surrounding Facebook's impact
on our families and communities.
I do not feel qualified to speak to the merits of the CAMRA Act
legislation. I do, however, support research into the effects of
technology and media, including social media and on children in their
cognitive, physical, and socio-emotional development. I support this
because those findings are essential to informing both oversight and
regulatory bodies as well as to media and social media companies who
are enabling access and exposure of these products to children. I
believe informed decision-making is essential.
I strongly encourage laws mandating that Facebook publicly disclose
a ``Firehose'' of content seen by those under the age of 18 in the
United States (or even by State, as the effects of Instagram are not
consistent across the country). If Facebook could be required to
publish 10 percent of all the public posts that were seen by more than
1000 people, along with appropriate metadata, we could confirm for
ourselves that Instagram is performing as the company publicly
represents. This is particularly true regarding children. The problems
on Facebook and Instagram are not about bad people or bad content--it
is about an algorithm-based platform that systematically promotes the
most extreme and divisive content.
Facebook has the ability to segment users of the platform based on
their interests and behavior in a manner that still protects personal
privacy e.g., by creating clusters with numbers rather than user
attributes. I myself worked on such a project in order to understand
the insidious and hard to detect practice of ``narrowcasting'' used in
targeted misinformation and disinformation campaigns.
The harms on Instagram are not evenly distributed across the
platform but rather disproportionately experienced by the most
vulnerable young people using it. Given the need to understand the
disparity of the problematic effects, it is important to understand
what content is being delivered to the median user versus the
vulnerable user. Being able to see differences in what content is
viewed by different segments of children will help us understand
whether and how some children are being pulled down harmful rabbit
holes.
Question 2. Another issue I'm very concerned about is the
protection of consumers' personal data, particularly when it comes to
data collected through biometric software such as facial recognition
technology. From your time working at Facebook, do you think the
company is doing enough to protect the privacy and security of its
users' data?
Answer. It is important to understand that there is more than one
type of personal data over which users may wish to protect their
privacy and security. There is personal data as publicly understood,
such as an individual's real identity and confidential personally
identifying information, that is overtly collected pursuant to the user
agreements required to launch an app on a person's device. Facebook
also collects and creates digital profiles of users and non-users
through the analysis of their behavior--e.g., usage history and
activities. This latter category can be much more invasive as it is
used for targeted advertising, marketing and platform performance,
growth strategies, micro-targeting as well as for predictive modelling.
I have no knowledge or information about protection over the latter
type of personal data.
I do think it should be legally mandated that Facebook publicly
disclose how personal data is used, including but not limited to how
ads are targeted and how group recommendations are selected, because
the sum of the usage of our personal data leads to radically different
outcomes.
______
Response to Written Questions Submitted by Hon. John Thune to
Frances Haugen
Question 1. Ms. Haugen, the public's trust of big tech platforms
has been significantly diminished over the past few years and
rightfully so. I believe we should consider providing incentives for
whistleblowers in the technology sector to shed light on some of the
practices of big tech platforms to Federal regulators. One of the
provisions of the PACT Act would require an assessment of establishing
a whistleblower protection and award program for employees who work for
social media platforms. What are the challenges you've faced as a
whistleblower that you think Congress could effectively address?
Answer. Thank you, Senator Thune, for your excellent questions and
your care and consideration of the issues surrounding the incentives
Facebook faces and ensuring the appropriate protections for
whistleblowers, such as myself, that are able to bring essential
information and knowledge into the public discourse and to the
attention of policy-makers, oversight and regulatory bodies.
Addressing challenges faced by whistleblowers is absolutely
critical in my opinion.
In my own case, when I became alarmed by the information I had
while working inside Facebook, I had no knowledge of any way in which I
could make my concerns known to anyone who could reliably or
independently consider them--let alone act to address these very
serious concerns. As I have said before, if I drove a bus or worked in
a hospital, there would be a phone number in my break room saying,
``Did you see something that endangers public safety? Please call this
number and someone will take you seriously''. There is no such
mechanism for employees in social media platforms writ large across the
industry--and it imposes a high personal cost on the employees that
work there because unless potential whistleblowers have the luxury of
being able to risk their jobs or careers, they are forced to live with
secrets that endanger people's lives. Even if some companies were to
have internal mechanisms for reporting problematic behavior, in my
experience, it is unlikely that Facebook employees such as myself would
trust that mechanism given Facebook's track record of ignoring the
public good.
It is a very difficult decision to come forward as a
whistleblower--especially when working in a powerful global corporation
with seemingly limitless influence. I struggled for a long time to find
a way forward because I knew that the information needed to be in the
hands of someone who could assess and act upon what was happening
inside the company. This very information was actively being hidden
from the public, the users, and the oversight and regulatory bodies.
Further, from my first day of orientation, I witnessed the company
create a culture against raising concerns with anyone lest I alienate
my ``Facebook family'' and be labelled a traitor. In my experience,
this culture of blind loyalty was pervasive and intimidating to the
point that I felt I must follow the company line to succeed inside
Facebook.
I know there are many people inside social media companies that
have this same experience of helplessness--particularly because there
are numerous examples of tech whistleblowers whose lives are destroyed
after bringing forward their concerns. It is unacceptable in my opinion
that there is no mechanism to safely report a concern which has
potentially massive adverse effects on even one individual user.
I support any actions that will lead to safe avenues for
whistleblowing. While I am very thankful for the SEC protections
provided by the Sarbanes-Oxley Act of 2002 and the Dodd-Frank Act of
2010, I remained concerned about the lack of awareness of such
protections for former employees like me, as well as concern about the
seeming lack of protections for whistleblowers in non-publicly traded
companies which may have similar implications for user safety and
security. At a minimum, we should extend the current whistleblower
protections to private companies. Ideally, all employers and employees
will be properly trained concerning the mechanisms by which to make
lawful disclosures and the protections afforded to whistleblowers who
do so.
Lastly, social media companies specifically should be held to a
higher standard because they have the ability to shape our perception
of reality. The scope of the danger of harm should be a factor in
required standards of social responsibility. Due to the closed nature
of these companies, whistleblowers are even more important, as there is
no other mechanism for getting clear information out of these walled-
off systems. As technology continues to accelerate faster and faster,
whistleblowers will become an essential counterbalancing mechanism for
society, as otherwise technology will be far further down the road
before we realize what externalities it is exerting on the public and
the consequences of those externalities.
It is important to remember that a whistleblower who comes forward
today must weigh the public good against losing a secure future that
can only be maintained by remaining silent. I strongly encourage the
establishment of reward programs with whistleblower protections to
offset some of this risk and to provide a way to acknowledge the public
service these whistleblowers do.
Question 2. Facebook has stated that as part of its efforts to make
its platform a healthier place for kids and teenagers, that it was
looking at adding ``nudges'' to its platform. The ``nudges'' feature
would be designed to nudge someone viewing potentially harmful content
towards something more positive. How confident are you that this latest
feature by Facebook will improve young users' experience?
Answer. I have no personal knowledge or information about the
specific ``nudges'' referenced in this question. However, based on my
experience at Facebook and the evidence I have disclosed to the SEC and
to the U.S. Congress, Facebook has a track record of publicizing
efforts to make their platforms safer, though in reality there may be
little improvement to diminishing harms facing users on any of
Facebook's Family of Apps. In the disclosed documents, there is an
example of this behavior in the case of ``Project Daisy''. Project
Daisy became a PR-friendly project announcement. Facebook management
decided to implement the removal of the ``like'' button in order to
appeal to the press and the public despite the research findings that
such a change would not materially reduce the harm to users. Facebook
did this again with their eating disorder and self harm intervention--
they claimed it was a clear indication of the commitment of the company
to protecting kids, yet it only triggers hundreds of times per day
rather than at a meaningful pace, because it is so narrowly tuned.
Other examples are the frequent public announcements of what
content will no longer be permitted on the platform. In these
instances, Facebook will announce that particular speech will not be
acceptable on its platform but fails to disclose that prohibited
content classifiers are used oftentimes because internally it is known
that this approach is woefully inadequate. By their own research
Facebook knows that content classifiers are largely ineffective or, for
most languages aside from English, non-existent.
This is why Facebook must be required to publish which AI labeling
systems they have, which languages they cover, and the effectiveness of
those systems--language by language. The United States is a
linguistically diverse country, and the public must be aware that
people speaking languages beyond English will be exposed to more
extreme, polarizing, and divisive content on Facebook-potentially
leading to radicalization. Facebook has said and will say again that
they can't tell you which safety systems they have because ``bad guys''
will find out and that will be dangerous to our country. But the
reality is the `bad guys' already know where Facebook's holes are
because they are continuously testing where Facebook is weak. The ``bad
guys'' are experts at exploiting the vulnerabilities of the platform
which is a separate problem without adequate resources in combating
influence operations and inauthentic behavior. Right now, based on my
experience and understanding, the only people who are unaware of how
deficient Facebook is are the ``good guys'' in places like the U.S.
Congress, or other oversight, regulatory or law-enforcement bodies.
It was clear during my experience at Facebook, through the
corporate structure and through the many experiences of my colleagues,
that decisions about public statements or announcements of positive
actions taken by the company are based more on anticipated receptivity
by the press and the public than actual positive impact of the
announced action.
Proposals to introduce ``nudges'' often make me nervous, because I
don't believe Facebook should be allowed to define what is ``good'' or
``bad'' content without substantial transparency and oversight from the
public. However, I do think that identifying kids who are viewing
increasing amounts of self harm content or eating disorder material and
providing interventions is important.
As you consider this and other Facebook-proffered mitigation
actions, please note that because of Facebook's history of secrecy and
misleading public assurances, today we have no idea what topics
Facebook believes ``nudges'' are needed for, how they define those
topics, or the effectiveness of the systems for identifying this
content. If it is anything like hate speech, Facebook's internal
research says that only 3-5 percent of that content will be identified,
with a future maximum ceiling of 10-20 percent effectiveness achieved
only with substantial further investment. This is not an effective
model.
Lastly, Facebook has a pattern of behavior of waiting until the pot
is boiling over before intervening, rather than slowly turning down the
temperature as the water warms. If the solution is based on content (as
opposed to making the platform slower and more human-scale over all), I
would anticipate that they would wait until there is a public
perception that kids are being significantly harmed before taking and
intervening actions. Substantial and unnecessary harm to children would
result from such a strategy.
Question 3. The PACT Act would promote transparency by requiring
Internet platforms like Facebook to disclose in detail their moderation
and censorship practices, and give more due process to users whose
content has been taken down. Do you believe this provision would help
build trust with Facebook's users?
Answer. Transparency is essential for public trust. Given Facebook
has made repeated assurances about improvements to the content
moderation process, such as not tolerating various types of speech
while harmful content continues to be amplified on the platform, they
should be required to provide detailed methodology on HOW they execute
each of their public promises.
It is important for Facebook to disclose exactly which content
labeling systems (e.g., hate speech, self harm, nudity, etc.) exist,
which languages they exist in (including dialects), and the efficacy of
those systems in each language (ex. precision and recall). I believe
Facebook has all of this information already and tracks these data
points internally. Right now there are widely divergent outcomes for
moderation and censorship on the platform. For example: I would be
unsurprised if speech that would be taken down in English was allowed
to stay up in Chinese, because Chinese language issues have had
radically less investment by Facebook than those in English. This is
not disproportionate to the risks of harm and it is dangerous as it has
resulted in, and will continue to impact, radicalization and
susceptibility to disinformation campaigns.
I believe that Facebook should have to disclose samples of real
content that meet each percentile of score within their systems (e.g.,
something that has a 50th percentile score looks like X and a 90th
percentile score looks like Y), and disclose at which scores do they
either take down or demote content. We have seen in Facebook's own
research that counter-terrorism speech is often misclassified as
terrorism-promoting speech due to the limitations of how content-based
enforcement systems operate and Facebook's chronic under-investment in
these systems. If the public could see a thousand examples once a week
of each score percentile, independent analysts would be able to ensure
that critical public safety information like counter-terrorism speech
is not erroneously removed from the platform. This is an example of how
Facebook's over-reliance on AI versus making the platform focus on
family and friends makes the platform more dangerous. Counter-terrorism
speech by people you know is the most effective way to prevent
radicalization and the rise of extremism.
Lastly, Facebook must publish the code used to produce any
information disclosed to the public. I cannot emphasize this point
enough. As a data scientist, I am aware of how every data analysis
requires assumptions and simplifications. Facebook has repeatedly
demonstrated in the past that Facebook cannot be trusted to disclose
the data they claim they are disclosing because they manipulate data to
serve their needs. They will put in assumptions invisibly behind the
scenes that benefit them or support their public narratives--which hide
the truth. If Facebook is required to publish how their data is
produced, independent researchers can explain and verify what they
claim their data is saying is what is actually occurring on the
platform.
The path to trust is earned, and meaningfully increasing
transparency is vital to rebuilding Facebook's relationship with the
public.
Question 4. As you have testified, an algorithm is not benign code
that is developed in a vacuum. An algorithm is a piece of software
designed by software engineers in consultation with lawyers and policy
experts.
Please describe in as much detail as you are able the internal
processes by which Facebook designs, deploys, and updates its content
selection algorithms, and provide citations to relevant internal
documents in your possession that detail these processes.
Answer. It is important to note that algorithms do not exist in
isolation from other product changes. For example, having Groups with 1
million members that can inject content into your news feed in the
absence of any agreement from you is a product choice that
substantially increases extreme and polarizing information in a user's
feed. At the same time, this is only possible because of how Facebook's
algorithms are designed. For example, if Facebook designed Groups to
work using Chronological ranking, they could not inject content into
your feed because it would be you rather than the AI focusing your
attention.
At a high level, there are two kinds of ranking changes for social
media platforms--1) changes designed to improve a goal metric or
metrics, and 2) integrity/safety changes that remove content or change
the position of content in the feed.
At Facebook, some algorithm change ideas are instigated by product
managers to achieve specific performance goals whether to increase goal
metrics like Meaningful Social Interactions (MSI). Other times they are
implemented to reduce harm by manipulating factors in the main scoring
function or by creating rate limits. These changes are researched and
tested by software engineers whose findings may become recommendations
for changes to the core algorithm. Once experiments have been run, the
findings and recommendations are sent to other teams inside Facebook
for consideration, including the Growth and the Policy team. Decisions
on whether to implement changes require agreement and approval by these
and other groups in order to be ``turned on'' or implemented.
Similarly, the impact of algorithm changes are closely monitored after
launch and may be recommended to be ``turned off'' by various teams
inside Facebook if they hurt core metrics like the # of sessions or
length of sessions.
There are multiple documents within my disclosure that discuss how
the process of improving algorithms at Facebook is poorly understood.
Scores are not calibrated, so a piece of content might have a score
from -5,000 to +100,000. As a result, safety related changes often are
undone by people working on growth related changes--the right hand does
not talk to the left hand, and because algorithms aren't easy to
understand, safety changes regularly regress.
Question 5. In 2014, a peer reviewed article appeared in the
Proceedings of the National Academy of Sciences entitled ``Experimental
evidence of massive-scale emotional contagion through social
networks.'' The article revealed that Facebook had conducted a massive
experiment on its platform that found that ``emotional states can be
transferred to others via emotional contagion, leading people to
experience the same emotions without their awareness.''
To your knowledge, has Facebook conducted any other studies like
these for its internal use? If so, please provide as much detail and
internal documents and communications about these studies as you are
aware.
Answer. I'm not aware of research related to this. I do know that
engagement-based ranking will pull people together into rabbit holes
where they reinforce the emotional state of each other. Many have said
``the solution to bad speech is more speech''--unfortunately within a
rabbit hole and whatever extreme beliefs exist there, people who
question extreme content will be silenced by the group via comment
pile-ons or by being removed from the group. In my experience, within
groups that are united by a belief in extreme content, there is no
mechanism for reporting problematic content, rendering intervention
based on reporting ineffective.
Question 6. You have worked closely with the Wall Street Journal, a
News Corp publication, to reveal thousands of internal Facebook
documents. You have testified that you were the lead Product Manager on
the Civic Misinformation team at Facebook, which dealt with issues
related to democracy and misinformation during the 2020 presidential
election. As you know, on October 14, 2020, Andy Stone, Policy
Communications Director at Facebook, tweeted that Facebook would be
``reducing its distribution'' of an article dated that same day
headlined ``Smoking-gun e-mail reveals how Hunter Biden introduced
Ukrainian businessman to VP dad,'' published in another News Corp
publication, the New York Post. Stone further tweeted that ``this is
part of our standard process to reduce the spread of misinformation.''
However, Ben Schreckinger, a reporter at Politico, has independently
confirmed that this e-mail and others uncovered by the New York Post
were, in fact, authentic. Clearly, Facebook harmed our democracy by
suppressing true information about the corruption of a presidential
candidate at the height of the 2020 presidential campaign.
Please describe in detail the decision-making process at Facebook
when it decided to limit distribution of this New York Post article on
its platform.
Answer. I have no information or knowledge about the post or the
decision referenced above. While the story referenced likely went
through Facebook's process for determining the veracity of stories via
the third party fact checking program, I do not know for sure.
Question 7. Did you have a hand in any of the decision-making at
Facebook to suppress this New York Post article? If so, please explain
in detail your role in suppressing this article. Do you have any
internal documents or communications in your possession regarding
Facebook's decision to suppress this New York Post article? If so,
please describe these documents and communications in detail.
Answer. To be clear--my team, Civic Misinformation, was responsible
for misinformation in places that did not have third-party fact
checking (i.e., most of the world/most languages in the world). The
team that would have been tasked with assessing the accuracy of that
article was the Viral Misinformation team (also known as just the
Misinformation team). This team also did not have regular interactions
with the Civic Integrity group, since it was within the larger
Community Integrity organization.
I worked on the Civil Integrity and the Threat Intelligence teams
while I was at Facebook. As such, I did not have any direct role in or
knowledge of Facebook's decisions or actions related to issues of
moderating U.S. focused content.
Question 8. What are the lessons learned from this specific
experience, in your view?
Answer. Without direct knowledge or involvement in this decision, I
cannot speak to what considerations were made nor what lessons were
learned from the experience. Nonetheless, the internal documents make
clear that the company's level of transparency to date has been
inadequate and problematic.
I am a strong proponent of non-content based solutions because they
are the only solutions that scale to linguistically diverse places, and
they reduce the chance of bias on the part of Facebook on which content
is actioned. Non-content based solutions are things like requiring
someone to click on a link before they reshare it, or limiting chains
of reshares to a depth of two (ex. Alice posts, Bob reshares, Carole
reshares, now if Dan wants to reshare it, he has to copy/paste). In the
case of limiting reshares to depth of 2, it has a similar impact on
misinformation to the entire third party fact checking program.
______
Response to Written Question Submitted by Hon. Marsha Blackburn to
Frances Haugen
Question. Are you aware of any other specific instances of human
exploitation content that Facebook failed to remediate once identified?
If so, why did they fail to take action (to the best of your
knowledge)?
Answer. Thank you, Ranking Member Blackburn, for your care of and
commitment to these issues as well as Facebook's impact on our families
and communities. In my experience, human exploitation content and other
extremely repugnant use of the platform were not given the resources
necessary to materially address the problems at Facebook.
When I left Facebook, the threat investigator team dedicated to
finding networks of Human Exploitation was composed of less than 10
employees--this team covered topics as broad as human trafficking, sex
trafficking, trafficking in organs, child trafficking and labor
trafficking, for every language across the platform. This illustrates
that Facebook is aware that these criminal networks exist on the
platform. Many journalists have written articles since the WSJ broke
this story demonstrating it is trivially easy to find ads on the
platform for buying and selling human beings.
The reality is that because Facebook does not have to disclose
publicly which threat investigator teams exist, and exactly how many
people work on each problem space and in each role, Facebook never has
to invest an appropriate number of people to stop human trafficking.
The documents show that Facebook took targeted action to address the
concerns raised on human trafficking on the platform in 2019 only once
Apple issued a letter threatening to remove the Facebook family of Apps
(``FOA'').
Facebook has taken a higher level responsibility because they hide
information on how their platform works from the public. As a result,
Facebook undertakes critical national security actions (like fighting
terrorism, cartels, human trafficking networks) in a vacuum, because
Facebook will not allow anyone outside the company to work with its
data in this effort. When our country is subjected to targeted and
pervasive information and influence campaigns conducted by hostile
adversaries including nation-states, I believe this has created a very
dangerous vulnerability to our national security. In contrast, Twitter
allows security researchers to supervise 1/10th of all the public
tweets--a very different shared responsibility for safety. If Facebook
is going to insist that they alone can work on stopping human
trafficking on the platform, they must be forced to disclose in detail
all the steps being taken so that Congress can evaluate if those
actions are indeed sufficient and proportional to the dangers that
exist.
[all]