[Congressional Record Volume 170, Number 125 (Wednesday, July 31, 2024)]
[Senate]
[Pages S5655-S5656]
From the Congressional Record Online through the Government Publishing Office [www.gpo.gov]



                   Unanimous Consent Request--S. 3875

  Madam President, there is a second bill which is S. 3875, that 
Senator Murkowski and I have. And, again, these two bills can mesh 
together. This is a bill that requires disclaimers on political ads 
substantially generated by AI. And I note ``substantially generated by 
AI.'' This is not about changing a hair color or doing a minor thing.
  While we must ban the most deceptive deepfakes, as I have just 
described, in our elections, it is also critical that voters know if 
ads they are seeing are made with this technology. This would 
especially help in cases of parody; in cases where, for instance, the 
video that was recently posted this last weekend by Elon Musk--which is 
a lengthy video--which takes the voice of candidate Kamala Harris, Vice 
President Kamala Harris, and puts her exact voice into words and 
sentences that she did not say. And while it is a parody and it 
wouldn't fall under the deepfake ban, it should require--even by X's 
own rules--a disclaimer

[[Page S5656]]

placed on this video; yet there was no disclaimer. And I am very afraid 
that if we are going to allow this stuff--believe me, some people see 
that and they need to be told it is generated by AI because when they 
only watch a few sentences of it and they actually think it is her 
saying these things--which, of course, it wasn't--because they piece 
together and scrape together her voice to say things that she didn't 
say, they are not going to know what it is.
  And I talked to colleagues on both sides of the aisle who have seen 
these kind of things that are done with some humor so they don't make 
the cut for the deepfakes, but they believe that they should say that 
it is prepared by AI so that people at least realize it is not the real 
voice of the candidate they like or the candidate they don't like.
  This bill, the bill with Senator Murkowski, is about making sure that 
voters can make their own decisions about what they are seeing and 
hearing and how it is being used to influence their vote.
  It is on solid Constitutional ground with the Supreme Court having 
repeatedly upheld disclosure laws. I just don't think, in the world, 
you are going to be able to say that this isn't Constitutional when the 
Supreme Court has held up these disclosure laws, and it simply gives 
our citizenry a way to evaluate whether or not that is a candidate's 
real voice or not.
  There are days where--I cannot even believe I am saying this--when 
all of these conservative States like Mississippi have actually put 
these laws into place for their own State political advertising. But in 
this Chamber, when it comes to Federal candidates for the congressional 
seats and the Senate and the Presidency, we have just decided: Nope, we 
are just going to let this go. Let's see what happens. Let's not know 
if our citizens are going to understand if it is us or not. We are not 
even going to give them the courtesy of letting them know with a 
disclaimer that it is done with AI.
  This bill incorporates feedback that we heard at a Rules Committee 
hearing by making clear that it does not apply when AI is used in minor 
ways, like for cosmetic adjustments, color editing, cropping.
  Of the 18 States that have passed the laws to regulate AI in 
election, 8 States--across the political spectrum, as I noted, 
including Utah which passed this law unanimously--have enacted laws to 
require disclaimers for AI-generated political ads.
  That would include the State of Florida. The State of Florida has put 
this in place. I never thought I would say that the State of Florida 
was more ahead of the Federal Government when it came to making sure 
that at least their citizens understood what they were seeing when they 
watched an ad.
  Indiana, Idaho, New York, Oregon, Wisconsin, and Washington have all 
passed similar laws to this one.
  So unless we are going to claim those laws, that DeSantis signed a 
law that was unconstitutional--my colleagues, if they want to claim 
that, I don't believe for a minute it is unconstitutional.
  And while some tech companies now have policies to require 
disclaimers on ads like this, with this year's election approaching, we 
need a consistent standard. That is why this bill was endorsed by the 
same group of over 40 national security experts and current and former 
senior government officials on both sides of the aisle.
  For these reasons, I urge my colleagues to join me in supporting this 
bipartisan measure to increase transparency in our elections and ensure 
voters are informed as they cast their ballot later this year.
  Senator Fischer and I have worked together very well, chair and 
ranking member of this committee; and I am still hopeful that, at least 
for this bill, when we come back in the fall, that we will be able to 
work something out so at least disclaimers are required.
  Madam President, as if in legislative session and notwithstanding 
rule XXII, I ask unanimous consent that the Senate proceed to the 
immediate consideration of Calendar No. 389, S. 3875; further, that the 
committee-reported substitute amendment be agreed to; the bill, as 
amended, be considered read a third time and passed; and the motion to 
reconsider be considered made and laid upon the table with no 
intervening action or debate.
  The PRESIDING OFFICER. Is there objection?
  The Senator from Nebraska.
  Mrs. FISCHER. Madam President, reserving the right to object, the AI 
Transparency in Elections Act echoes the Honest Ads Act and the 
DISCLOSE Act.
  Those bills failed to become law because they created new Federal 
burdens on the foundational right of Americans to free speech. Adding a 
new definition of AI to these partisan bills does not resolve these 
concerns.
  I would welcome a thoughtful policy proposal to address the actual 
concerns posed by AI-generated deepfakes. Instead, my colleagues are 
attempting to recycle an already failed proposal, and, therefore, I 
object.
  The PRESIDING OFFICER. Objection is heard.
  The Senator from Minnesota.
  Ms. KLOBUCHAR. Mr. President, with this year's election now in less 
than 100 days, we must put in place commonsense rules of the road to 
address the risk that AI poses for our democracy.
  The risks are clear: We have heard repeatedly about the potential for 
AI to upend our elections. All of our witnesses, from both parties, 
agreed that this was a threat when we had our Rules Committee hearing.
  And at the bipartisan AI forum, led by Leader Schumer, with Senators 
Rounds, Heinrich, and Young, we heard consensus that Federal 
legislation is necessary; that disclaimers are necessary for certain 
ads. And for others, we simply must take them down when they are 
pretending to be a candidate and deliberately and intentionally 
misleading voters. I don't care what party does it. I don't care what 
super PAC does it. There is absolutely no way--and these State 
legislatures have agreed with me, nearly unanimously, that we shouldn't 
at least have a disclaimer on them, much less to ban them, which is 
what Senators Hawley and Collins and Coons and I are suggesting in the 
deepfake bill.
  This is a hair-on-fire moment. AI has the potential to turbocharge 
the spread of disinformation and deceive voters. This is why we must 
take action. And I hope that when people see what is going on through 
August, when sometimes early voting has started, we will give at least 
the tools to the platforms to be able to point to a Federal law--most 
of these other State laws have just been adopted in the last few 
months--and say this is not OK; that they have a right to at least 
require a disclaimer on these ads, just like they do on TV for various 
things--and we have all seen it--so that we know what is going on.
  Democracy dies in the darkness, as one newspaper has said. And we are 
literally putting a veil over people's faces if we are not allowing 
them to assess whether or not the person is really the person that they 
are looking at on their phone or hearing in a robocall.
  I just think it is outrageous if we let this continue. And I 
appreciate that there are Republican Governors in States and Republican 
legislatures who have actually seen this as I see it and that there is 
bipartisan support for this in the U.S. Senate. And I hope that in the 
fall we will revisit this.
  The PRESIDING OFFICER. The Republican whip.