[House Hearing, 116 Congress] [From the U.S. Government Publishing Office] INCLUSION IN TECH: HOW DIVERSITY BENEFITS ALL AMERICANS ======================================================================= HEARING BEFORE THE SUBCOMMITTEE ON CONSUMER PROTECTION AND COMMERCE OF THE COMMITTEE ON ENERGY AND COMMERCE HOUSE OF REPRESENTATIVES ONE HUNDRED SIXTEENTH CONGRESS FIRST SESSION __________ MARCH 6, 2019 __________ Serial No. 116-13 [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT] Printed for the use of the Committee on Energy and Commerce govinfo.gov/committee/house-energy energycommerce.house.gov _________ U.S. GOVERNMENT PUBLISHING OFFICE 37-565 PDF WASHINGTON : 2020 COMMITTEE ON ENERGY AND COMMERCE FRANK PALLONE, Jr., New Jersey Chairman BOBBY L. RUSH, Illinois GREG WALDEN, Oregon ANNA G. ESHOO, California Ranking Member ELIOT L. ENGEL, New York FRED UPTON, Michigan DIANA DeGETTE, Colorado JOHN SHIMKUS, Illinois MIKE DOYLE, Pennsylvania MICHAEL C. BURGESS, Texas JAN SCHAKOWSKY, Illinois STEVE SCALISE, Louisiana G. K. BUTTERFIELD, North Carolina ROBERT E. LATTA, Ohio DORIS O. MATSUI, California CATHY McMORRIS RODGERS, Washington KATHY CASTOR, Florida BRETT GUTHRIE, Kentucky JOHN P. SARBANES, Maryland PETE OLSON, Texas JERRY McNERNEY, California DAVID B. McKINLEY, West Virginia PETER WELCH, Vermont ADAM KINZINGER, Illinois BEN RAY LUJAN, New Mexico H. MORGAN GRIFFITH, Virginia PAUL TONKO, New York GUS M. BILIRAKIS, Florida YVETTE D. CLARKE, New York, Vice BILL JOHNSON, Ohio Chair BILLY LONG, Missouri DAVID LOEBSACK, Iowa LARRY BUCSHON, Indiana KURT SCHRADER, Oregon BILL FLORES, Texas JOSEPH P. KENNEDY III, SUSAN W. BROOKS, Indiana Massachusetts MARKWAYNE MULLIN, Oklahoma TONY CARDENAS, California RICHARD HUDSON, North Carolina RAUL RUIZ, California TIM WALBERG, Michigan SCOTT H. PETERS, California EARL L. ``BUDDY'' CARTER, Georgia DEBBIE DINGELL, Michigan JEFF DUNCAN, South Carolina MARC A. VEASEY, Texas GREG GIANFORTE, Montana ANN M. KUSTER, New Hampshire ROBIN L. KELLY, Illinois NANETTE DIAZ BARRAGAN, California A. DONALD McEACHIN, Virginia LISA BLUNT ROCHESTER, Delaware DARREN SOTO, Florida TOM O'HALLERAN, Arizona ------ Professional Staff JEFFREY C. CARROLL, Staff Director TIFFANY GUARASCIO, Deputy Staff Director MIKE BLOOMQUIST, Minority Staff Director Subcommittee on Consumer Protection and Commerce JAN SCHAKOWSKY, Illinois Chairwoman KATHY CASTOR, Florida CATHY McMORRIS RODGERS, Washington MARC A. VEASEY, Texas Ranking Member ROBIN L. KELLY, Illinois FRED UPTON, Michigan TOM O'HALLERAN, Arizona MICHAEL C. BURGESS, Texas BEN RAY LUJAN, New Mexico ROBERT E. LATTA, Ohio TONY CARDENAS, California, Vice BRETT GUTHRIE, Kentucky Chair LARRY BUCSHON, Indiana LISA BLUNT ROCHESTER, Delaware RICHARD HUDSON, North Carolina DARREN SOTO, Florida EARL L. ``BUDDY'' CARTER, Georgia BOBBY L. RUSH, Illinois GREG GIANFORTE, Montana DORIS O. MATSUI, California GREG WALDEN, Oregon (ex officio) JERRY McNERNEY, California DEBBIE DINGELL, Michigan FRANK PALLONE, Jr., New Jersey (ex officio) (ii) C O N T E N T S ---------- Page Hon. Jan Schakowsky, a Representative in Congress from the State of Illinois, opening statement................................. 1 Prepared statement........................................... 2 Hon. Tony Cardenas, a Representative in Congress from the State of California, prepared statement.............................. 3 Hon. Robin L. Kelly, a Representative in Congress from the State of Illinois, prepared statement................................ 4 Hon. Cathy McMorris Rodgers, a Representative in Congress from the State of Washington, opening statement..................... 5 Prepared statement........................................... 6 Hon. Frank Pallone, Jr., a Representative in Congress from the State of New Jersey, opening statement......................... 7 Prepared statement........................................... 8 Hon. Bobby L. Rush, a Representative in Congress from the State of Illinois, prepared statement................................ 10 Hon. Richard Hudson, a Representative in Congress from the State of North Carolina, opening statement........................... 10 Witnesses Mark S. Luckie, Digital Media Strategist and Former Manager, Twitter and Facebook........................................... 12 Prepared statement........................................... 15 Answers to submitted questions............................... 150 Joan Ferrini-Mundy, Ph.D., President, University of Maine........ 23 Prepared statement........................................... 25 Answers to submitted questions............................... 152 Jiny Kim, Vice President, Policy and Programs, Asian Americans Advancing Justice.............................................. 34 Prepared statement \1\....................................... 36 Nicol Turner Lee, Ph.D., Fellow, Center for Technology Innovation, Governance Studies, Brookings Institution.......... 43 Prepared statement........................................... 45 Answers to submitted questions............................... 154 Natalie Oliverio, Founder and Chief Executive Officer, Military Talent Partners................................................ 63 Prepared statement........................................... 64 Answers to submitted questions \2\........................... 156 Jill Houghton, President and Chief Executive Officer, Disability:IN.................................................. 68 Prepared statement \3\....................................... 70 Answers to submitted questions............................... 159 ---------- \1\ A February 2017 report entitled ``Breaking the Mold: Investing in Racial Diversity in Tech'' has been retained in committee files and also is available as part of Ms. Kim's written testimony at https://docs.house.gov/meetings/IF/IF17/ 20190306/108901/HHRG-116-IF17-Wstate-KimJ-20190306.pdf. \2\ Ms. Oliverio did not answer submitted questions for the record by the time of publication. \3\ A report entitled ``The 2018 Disability Equality Index: A Record Year for Corporate Disability Inclusion and Leadership'' has been retained in committee files and also is available as part of Ms. Houghton's written testimony at https:// docs.house.gov/meetings/IF/IF17/20190306/108901/HHRG-116-IF17- Wstate-HoughtonJ-20190306.pdf. David Lopez, Co-Dean, Rutgers Law School-Newark.................. 74 Prepared statement........................................... 76 Answers to submitted questions............................... 162 Submitted Material Article of July 26, 2018, ``Amazon's Facial Recognition Wrongly Identifies 28 Lawmakers, A.C.L.U. Says,'' by Natasha Singer, The New York Times, submitted by Mr. Rush...................... 124 Letter of March 6, 2019, from Marc H. Morial, President and Chief Executive Officer, National Urban League, to Ms. Schakowsky and Mrs. Rodgers, submitted by Ms. Schakowsky...................... 128 Letter of March 5, 2019, from Marc Rotenberg, President, and Caitriona Fitzgerald, Policy Director, Electronic Privacy Information Center, to Ms. Schakowsky and Mrs. Rodgers, submitted by Ms. Schakowsky.................................... 130 Letter of March 5, 2019, from Sean Perryman, Director of Diversity and Inclusion Policy & Counsel, Internet Association, to Ms. Schakowsky and Mrs. Rodgers, submitted by Ms. Schakowsky 134 Statement of Jennifer Huddleston, Research Fellow, Mercatus Center at George Mason University, March 6, 2019, submitted by Ms. Schakowsky................................................. 136 Letter of March 5, 2019, from Maxine Williams, Chief Diversity Officer, Facebook, Inc., to Mr. Pallone, et al., submitted by Ms. Schakowsky................................................. 140 Letter, undated, from Hon. Maxine Waters, a Representative in Congress from the State of California, to Mr. Pallone, submitted by Ms. Schakowsky.................................... 146 Blog post of February 27, 2018, ``Expanding Apprenticeship Program Across the Country to Hire more Veterans,'' by Paul Marchand, Executive Vice President, Human Resources, Charter Communications, submitted by Ms. Schakowsky.................... 148 INCLUSION IN TECH: HOW DIVERSITY BENEFITS ALL AMERICANS ---------- WEDNESDAY, MARCH 6, 2019 House of Representatives, Subcommittee on Consumer Protection and Commerce, Committee on Energy and Commerce, Washington, DC. The subcommittee met, pursuant to call, at 10:33 a.m., in the John D. Dingell Room 2123, Rayburn House Office Building, Hon. Jan Schakowsky (chairwoman of the subcommittee) presiding. Members present: Representatives Schakowsky, Castor, Veasey, Kelly, O'Halleran, Lujan, Cardenas, Blunt Rochester, Soto, Rush, Matsui, McNerney, Dingell, Pallone (ex officio), Rodgers (subcommittee ranking member), Latta, Guthrie, Bucshon, Hudson, Carter, and Gianforte. Also present: Representatives Butterfield and Clarke. Staff present: Jeffrey C. Carroll, Staff Director; Evan Gilbert, Press Assistant; Lisa Goldman, Counsel; Waverly Gordon, Deputy Chief Counsel; Alex Hoehn-Saric, Chief Counsel, Communications and Technology; Meghan Mullon, Staff Assistant; Joe Orlando, Staff Assistant; Tim Robinson, Chief Counsel; Chloe Rodriguez, Policy Analyst; Melissa Froelich, Minority Chief Counsel, Consumer Protection and Commerce; Peter Kielty, Minority General Counsel; Bijan Koohmaraie, Minority Counsel, Consumer Protection and Commerce; Brannon Rains, Minority Staff Assistant; and Nate Wilkins, Minority Fellow. Ms. Schakowsky. The Subcommittee on Consumer Protection and Commerce will now come to order. I am going to say good morning, and thank you all for joining us today. And I will recognize myself first for 5 minutes with an opening statement. OPENING STATEMENT OF HON. JAN SCHAKOWSKY, A REPRESENTATIVE IN CONGRESS FROM THE STATE OF ILLINOIS Once again, good morning to everybody. Thank you to our witnesses. Today, we are meeting to discuss an important issue, the lack of diversity in the tech workforce. As the presence of technology continues to play a larger and larger role in all of our lives, industry's workforce has remained largely homogeneous. People of color, women, and older Americans have all been notably absent from the tech workforce, and the corresponding problem that that creates has been that the technology itself reflects that lack of diversity. And I want to hear about that today. This has real impact on Americans. We have seen algorithms biased against sentencing guidelines, resulting in harsher sentences for minorities. We have seen that automatic soap dispensers sometimes fail to recognize the hands of African Americans and Latinos--imagine that--who are seeking to use the product to wash their hands. These may seem like two opposite sides of the spectrum as far as harm, but they both clearly demonstrate that something is amiss. Simply put, diverse voices are lacking in the tech workforce. Moreover, diverse startups are facing difficulty competing with the large multinational technology companies. It strikes me that unfair business practices and extreme market concentration in tech may, in fact, perpetuate the bias and the old boys' club, or actually the young boys' club, that we are examining today. This lack of diversity in the workforce has real-life impact on consumers, and I thank our panel for coming here to discuss this very important issue. [The prepared statement of Ms. Schakowsky follows:] Prepared statement of Hon. Jan Schakowsky Good morning and thank you all for joining us today. Today, we are meeting to discuss an important issue, diversity in the tech workforce. As the presence of technology continues to play a larger and larger role in Americans' lives, the industry's workforce has remained largely homogenous. People of color, women, and older Americans have all been notably absent from the tech workforce. This has real impacts on Americans--we have seen algorithmic bias impact sentencing guidelines, resulting in harsher sentences for minorities. We have seen soap dispensers fail to recognize the hands of African Americans and Latinos seeking to use the product to wash their hands. These may seem like two opposite sides of the spectrum, as far as harm, but they both clearly demonstrate that something is amiss. Simply put, diverse voices are lacking in the tech workforce. It strikes me that unfair business practices and extreme market concentration in tech may in fact perpetuate the boys club that we are examining today. As previously stated, this lack of diversity in the workforce has real life impacts on consumers. I thank our panel for coming here to discuss this important issue. With that, I yield to my vice chair from California, Mr. Cardenas, for 1 minute. Ms. Schakowsky. With that, I want to yield 1 minute to the vice chair of this subcommittee, from California, Mr. Cardenas, for 1 minute. Mr. Cardenas. Thank you very much, Madam Chairwoman. It is no secret that the tech industry has a diversity problem, and every day we are seeing more and more unintended consequences when companies lack a diverse body of employees. We are seeing fitness trackers, for example, that have problems with dark skin. They just don't operate properly. Virtual assistants like Alexa having a hard time recognizing accents. Something that is no secret is that we still have something that is news to some people: Diversity is actually good for business. The Hispanic community in America has a buying power annually of upwards of $1.5 trillion. Hispanics make the fastest-growing number over number of growing Americans in this country. Also, it is a younger population, and Hispanics happen to have a high brand loyalty. So, it is good for business to have diversity, especially when it comes to Hispanics. Reports show that companies with more diversity amongst senior executives were 33 percent more likely to see an increase in their bottom line. When you have diverse backgrounds and experiences among your employees, you spur innovation; you avoid creating bias into your products; you avoid turning your back on a whole group of Americans. So, how do we solve this problem? I will say this: for example, a sharp Princeton-educated computer engineer recently told me she heard her coworker say that women and people of color dilute the talent pool for tech companies. I take that very personal. I am an electrical engineer myself. We have a problem in our culture in the tech industry that diversity is not only not prioritized, it's seen as a hindrance, of which nothing could be further from the truth. I would like to say much more, but, again, thank you, Madam Chairwoman, and I yield back the balance of my time. [The prepared statement of Mr. Cardenas follows:] Prepared statement of Hon. Tony Cardenas Thank you, Chairwoman Schakowsky. It's no secret that the tech industry has a diversity problem. Every day we're seeing more and more unintended consequences when companies lack a diverse body of employees. We're seeing fitness trackers for example that have problems with dark skin. Virtual assistants like Alex having a hard time recognizing accents. Something else that's no secret--but might still be news to some people--diversity is good for business. The Hispanic community in America has a buying power annually of upwards of $1.5 trillion. Hispanics have high brand loyalty. It's good for business to have diversity, especially when it comes to Hispanics. Reports show that companies with more diversity among senior executives were 33 percent more likely to see an increase in their bottom line. When you have diverse backgrounds and experiences among your employees, you spur innovation. You avoid accidentally embedding bias into your products. You avoid turning your back on a whole group of Americans. So how do we solve this problem? I'll say this--a sharp, Princeton-educated computer engineer recently told me she heard her coworker say that women and people of color dilute the talent pool for tech companies. Not only is that false--it's highly offensive. I myself am an engineer by training. We have a problematic culture in the tech industry when diversity is not only NOT prioritized--it's seen as a hinderance. And nothing can be further from the truth. Let's make technology work for all Americans and also help businesses succeed. I yield back my time to the chairwoman. Ms. Schakowsky. And I yield now the balance of my time to Congresswoman Kelly. Ms. Kelly. Thank you, Madam Chair. Thank you, Chairwoman Schakowsky and Ranking Member Rodgers, for holding this hearing today. As a founder of the Tech Accountability Caucus and founder of the Diversifying Tech Caucus with my colleague on the other side of the aisle, Cathy McMorris Rodgers, I am glad that this subcommittee is committed to addressing the issue of diversity in tech. Report after report from technology companies continue to show a lack of diversity in their workforces. According to the National Urban League, less than 5 percent of the digital workforce is African-American today. It is not just the large tech companies in Silicon Valley. Organizations like Mentoring Youth Through Technology, or MYTT, in my district help get minority students interested in STEM careers, but I continue to hear from startups in Chicago that they struggle to recruit diverse workforces. This is a fundamental problem, getting women and minorities into the technology jobs, and it must be corrected. A lack of diversity creates real-world problems of producing programs that can harm underserved communities. Poorly trained artificial intelligence tools can lead to implicit racial, gender, or ideological biases and can perpetuate existing biases. As AI use becomes more common and decisions are made by machines, we may not understand it is vital that these considerations are taken into account. I am hopeful that companies are going to see the benefits of having a diverse workforce that brings new ideas and perspectives. While there is no one solution to this problem, I hope that the witnesses today--and I am happy to see a former colleague, Dr. Turner Lee--will share their experiences and make recommendations, so we can continue to make tech a more diverse and inclusive community. Thank you, Madam Chair. [The prepared statement of Ms. Kelly follows:] Prepared statement of Hon. Robin L. Kelly Thank you, Chairwoman Schakowsky and Ranking Member Rodgers for holding this hearing today. As a founder of the Tech Accountability Caucus and founder of the Diversifying Tech Caucus with my friend on the other side of the aisle, Cathy McMorris Rodgers, I am glad that this subcommittee is committed to addressing the issue of diversity in tech. Report after report from technology companies continue to show a lack of diversity in their workforces. According to the National Urban League, less than 5 percent of the digital workforce is African-American today. It is not just the large tech companies in Silicon Valley. Organizations like Mentoring Youth Through Technology or MYTT in my District, help get minority students interested in STEM careers. But I continue to hear from start-ups in Chicago that they struggle to recruit diverse workforces. There is a fundamental problem getting women and minorities into the technology jobs and it must be corrected. A lack of diversity creates real world problems of producing programs that can harm underserved communities. Poorly trained artificial intelligence tools can lead to implicit racial, gender, or ideological biases and can perpetuate existing biases. As AI use becomes more common and decisions are made by machines we may not understand, it is vital that these considerations are taken into account. I am hopeful that companies are going to see the benefits of having a diverse workforce that brings new ideas and perspectives. While there is no one solution to this problem, I hope that the witnesses today will share their experiences and make recommendations so we can continue to make tech a more diverse and inclusive community. Ms. Schakowsky. Thank you. Thank you. And now, it is my pleasure to recognize for 5 minutes our ranking member, Ms. McMorris Rodgers. OPENING STATEMENT OF HON. CATHY McMORRIS RODGERS, A REPRESENTATIVE IN CONGRESS FROM THE STATE OF WASHINGTON Mrs. Rodgers. Thank you. Thank you, Madam Chair, and I appreciate you organizing this panel today, and appreciate the additional voices at the table. As my colleague Congresswoman Kelly just said, we organized the Diversifying Tech Caucus in 2015. It was aimed at highlighting some of these issues that we are going to hear more about today, getting more women, people with disabilities, minorities, veterans, into the tech sector. At a time when we are celebrating a booming economy because of our work on tax reform and to lift the regulatory burden, our economy has more jobs today. We have record low unemployment, record participation with African Americans, Hispanics, people with disabilities, people coming off the sidelines. It means that there are more opportunities for people to find better-paying jobs and have an opportunity for a better life. Today's hearing focuses, I believe, on an important aspect of encouraging those opportunities for all. And it is the focus on recruitment. We need to be doing more to recruit into the tech field, but also, once we recruit, to retain and promote these individuals into positions of leadership. The creation of one high-tech job is projected to create 4.3 other jobs in the local economy. I often say the job is the opportunity. It is vital that these opportunities are available to people from all walks of life. A vibrant and dynamic workplace with women, people of color, people with disabilities, reflects the promise of America, where, no matter who you are, you can achieve your version of the American dream. Oftentimes, it might look different than your own. I am excited personally for the opportunities for those with disabilities to work because more are offering the accommodations and the job coaches. Employing people with disabilities fosters innovation and it creates a stronger workplace culture. Many tech companies are leading in hiring those with autism. And because of their unique abilities for the attention to detail and the abilities to detect patterns, taking a software testing company like ULTRA Testing, the founder's wife one day told him, quote, ``We spend all this time focused on things these children may never be good at, but we spend no time nurturing the skills they already have a talent for. Isn't that a shame?'' And he agreed, and he got to work hiring people with autism to leverage their strengths. And now, ULTRA Testing is outperforming bigger companies in software quality assurance. This month, we are also celebrating Women's History Month, and it is a time to celebrate women who are leading across the board. Our goal today is to see more women in tech leading, being those disruptors, inspiring our next generation of transformational women leaders. Again, it may look different. Women have different leadership styles. And research is showing that we have greater understanding for teams and systems. We foster a healthy workplace culture. When women are not at the table, our perspectives and voices aren't represented. A study by McKinsey shows that companies with women in executive positions outperform the average profitability of their industries by 21 percent. Tech companies that don't open the door for women to shine and be decisionmakers risk being left behind. So, yes, we need to do more to open the door to allow these women to shine and do more to retain and promote these women. America has led the world in innovation. We celebrate that every single day. Entrepreneurs from all walks of life are taking an idea, making it a reality, creating more opportunities for hardworking people across the country. Again, that is the promise of America. It is not the promise for just some people or the somebodies in Silicon Valley. It is a promise for everyone. When we celebrate every person's strengths and abilities and embrace what every person has to offer, we are living up to that promise. I recognize there has been tremendous efforts, like recruiting more girls into STEM, and hiring people with disabilities, like ULTRA Testing that I mentioned. We need to continue to do more to address the pipeline, whether it is young people of every background, in girls in elementary and middle school, and exceptional people with disabilities, but we also need to focus on how we retain those individuals once they are recruited, and do more to encourage their promotion to leadership positions. So, today I look forward to hearing how the tech industry is leading on this and where you can also do better. Thank you all for being here. [The prepared statement of Ms. Rodgers follows:] Prepared statement of Hon. Cathy McMorris Rodgers Good morning. Today we are focused on an issue I have led on for quite some time: diversity in the tech industry. Today, we will have an opportunity to give credit where credit is due, while also exploring how we continue to do better.especially where we can continue to improve where we recruit, retain, and promote a more diverse workforce. I want to thank Chair Schakowsky for organizing this hearing today and for including all the voices at the table. I also would like to recognize my good friend from Illinois, Robin Kelly. In 2015, we launched the Diversifying Technology Caucus aimed at getting more women, people with disabilities, minorities, and veterans into the tech sector. I want to thank Ms. Kelly for working with me to promote more opportunities for all in the tech sector. This is an exciting time in America. Because of our work on tax reform and to lift the regulatory burden our economy is booming. After a decade of Americans asking, ``where are the jobs?'' wages are rising and there are more jobs available than people looking for work. As the Wall Street Journal just reported, women are driving the laborforce comeback. In addition, a record number of African Americans, Hispanics, and people with disabilities are coming off the sidelines and finding work. It means that more people are finding opportunities for a better life in healthcare, energy, construction, the service industry and more. Today's hearing is about ensuring more individuals have opportunities to pursue and advance careers in the tech industry too. The creation of one high tech job is projected to create 4.3 other jobs in a local economy. Because a job is the opportunity it's vital that these opportunities are available to people of all walks of life. A vibrant and dynamic workplace with women, people of color, people with disabilities, and more reflects the promise of America where no matter who you are, you can achieve your version of the American dream. Oftentimes it may look different and I'm excited about more opportunities for those with disabilities to work because of more commitments to accommodations and job coaches. Employing people with disabilities fosters innovation and it creates a stronger workplace culture. Many tech companies are leading in hiring people on the autism spectrum because of their unique abilities for attention to detail, and abilities to detect patterns. Take a software testing company called, ULTRA Testing. The founder's wife one day told him: ``We spend all this time focused on things these children may never be good at but we spend no time nurturing the skills they already have a talent for--isn't that a shame?'' He agreed, and got to work hiring people with autism to leverage their strengths. Now ULTRA Testing, a startup, is outperforming bigger companies in software quality assurance. This month we are also celebrating Women's History Month. It's a time to celebrate the women who are leading in tech being disruptors and inspiring our next generation of transformational women leaders. Again, it may look different. Women have different leadership styles. And research is showing that we have greater understanding for teams and systems and we foster a healthy workplace culture. When women are not leading at the table. our perspectives and our voices aren't represented. A study by McKinsey shows that companies with women in executive positions outperformed the average profitability of their industries by 21 percent. Tech companies that don't open the door for women to shine and be decision makers risk being left behind. Yes, that means hiring more women but it also means fostering an environment focused on retention and the promotion of women too. America is leading the world in innovation. Every single day, entrepreneurs from all walks of life are taking an idea making it a reality and creating more opportunities for hardworking people across the country. Again, that's the Promise of America. It's not the promise for just some people or the somebodies in Silicon Valley. It's a promise for everyone. When we celebrate every person's strengths and abilities and embrace what every person has to offer we are living up to that promise. I recognize there's been tremendous efforts like recruiting more girls into STEM and hiring people with disabilities, like at ULTRA Testing. We need to continue to do more to address the pipeline, whether it's young people of every background and girls in elementary and middle school. and exceptional people with disabilities. We also need to focus on how we retain those individuals once they are recruited and do more to encourage their promotion to leadership positions. So today, I look forward to hearing how the tech industry is leading on this and where you can also do better. Thank you to our witnesses. I yield back. Ms. Schakowsky. The gentlewoman yields back. And now, the Chair recognizes Mr. Pallone, chairman of the full committee, for 5 minutes for his opening statement. OPENING STATEMENT OF HON. FRANK PALLONE, Jr., A REPRESENTATIVE IN CONGRESS FROM THE STATE OF NEW JERSEY Mr. Pallone. Thank you, Madam Chair. As this subcommittee knows well, the influence of the internet and technology in our lives has grown exponentially over the past two decades, and our daily lives as consumers and workers have become dependent on technology. But while the U.S. has become more and more diverse, the workforce of the technology sector has not kept up. And we are seeing the effects of that in the products and services we use, like the wearable fitness trackers that don't work for people with dark skin, online job advertisements targeted at men over women, and websites with buttons and links too small for people with motor impairments to use. Without inclusive workforces, too often product design leaves people out, and the result can be embarrassing for the company when discovered and harmful for society when a discriminatory result is not identified and fixed. These are complicated and often uncomfortable discussions, but they are necessary to start to make changes. The Congressional Black Caucus launched its Tech 2020 Initiative in 2015 and has been working to hold companies accountable since. And several members of this committee have been working on these issues for years. I would like to yield time to some of them today. So, I will start beginning with Mr. Lujan. I would yield 1 minute to him. [The prepared statement of Mr. Pallone follows:] Prepared statement of Hon. Frank Pallone, Jr. As this subcommittee knows well, the influence of the Internet and technology in our lives has grown exponentially over the past two decades. Our daily lives as consumers and workers have become dependent on technology. But while the United States has become more and more diverse, the workforce of the technology sector has not kept up. And we are seeing the effects of that in the products and services we use-like wearable fitness trackers that don't work for people with dark skin, online job advertisements targeted at men over women, and websites with buttons and links too small for people with motor impairments to use. Without inclusive workforces, too often product design leaves people out. The result can be embarrassing for the company when discovered and harmful for society when a discriminatory result is not identified and fixed. These are complicated and often uncomfortable discussions, but they are necessary to start to make changes. The Congressional Black Caucus launched its Tech 2020 initiative in 2015 and has been working to hold companies accountable since. And several members of this committee have been working on these issues for years. I'd like to yield time to some of them today, beginning with Mr. Lujan for 1 minute. Thank you, I will also yield a minute to Mr. Butterfield for 1 minute. Thank you. And I will yield my final minute to Mr. Rush. Mr. Lujan. Thank you very much, Mr. Chairman. And I thank our Chair and our ranking member for this important hearing. When it comes to diversity in tech, let me be clear, more is needed. The tech industry is not where it needs to be on this issue. Representation of women and people of color in tech companies lags the rest of corporate America, and this matters. As Dr. Lee notes in her testimony, the absence of diversity among the people that make the decisions around products and services for the tech sector and the markets that these companies serve hurts us all. This lack of diversity informs the algorithms that determine whether people get a loan or a job, impacts how much people pay for everyday products and services. Investigations and studies have shown that these algorithms often have biased results and discriminatory outcomes. It is one reason why I am reintroducing the Accountable Capitalism Act with the inclusion of diversity language, to push corporations to make more socially responsible decisions. Action and accountability are a must. And I yield back. Mr. Pallone. And, Madam Chair, I would now like to yield a minute to Mr. Butterfield. Mr. Butterfield. Thank you, Chairman Pallone. Thank you, Chairwoman Schakowsky. For too long, the promise of the internet age for communities of color has been left unfulfilled. Technology is a fast-growing and lucrative industry, employing thousands every year. However, the rate of racial minorities in tech industries has not increased at a sufficient rate. This is unacceptable, Madam Chair, considering the number of capable racial minority students and workers that are available. The problem is exacerbated by insufficient resources in K- 12 schools to HBCUs and often low expectations for our minority students. Studies show that African-American children enter kindergarten at a competitive disadvantage. To close the gap in tech, we must give our students the resources they need to compete, not starting in college, but in the earliest stages of primary education. Public and private sectors have a responsibility to work together to create effective diversity and inclusion initiatives. The CBC Tech 2020 has been pounding tech companies now for years. We are beginning to see modest results. And so, I want to thank you for your effort in this space, and I yield back. Mr. Pallone. Thank you. Before I yield to Mr. Rush for the last minute, I just want to say, as you could probably tell, Mr. Rush had a subcommittee hearing in the Energy Subcommittee on diversity in the workforce. And now, we are doing this in the Consumer Protection Subcommittee with Ms. Schakowsky. Because the Energy and Commerce Committee has jurisdiction over so many sort of jobs of the future and areas of the future, I think it is particularly important that we address this issue of diversity. Mr. Rush, I yield the balance of the time. Mr. Rush. I want to thank you, Chairman Pallone, for yielding. And I want to thank you, Chairwoman Schakowsky, for holding this important hearing. In the 1990s, Chicago's O'Hare Airport, the world's busiest airport at the time, became the first to install touchless faucets. This innovation's promise of making things more sanitary and wasting less water were anticlimactic, however, when it was shown that these faucets had difficulties in recognizing the hands of black and brown Americans. While seemingly trivial, this is just one example of the real-world impediments that people of color across the Nation face because of their lack of representation in the technology industry. This same phenomenon has been repeated in facial recognition technology that mistakenly identified 28 Members of Congress, myself included, in search engines that provide ads related to criminal record history when associating, quote, ``black-sounding names,'' end of quote. Madam Chairman, it should be clear that, while the issue and emergency in tech may seem like a relatively straightforward problem, its ramifications are much bigger and go much deeper than it would appear. Madam Chairman, progress and innovation must go beyond being just skin deep. So, I thank you for holding this hearing and I look forward to hearing from the witnesses. And, Madam Chair, I would like unanimous consent to submit a New York Times article about mistakes in facial recognition for the record. And thank you. I yield back the balance of my time. Ms. Schakowsky. Without objection, so ordered. [The article appears at the conclusion of the hearing. The prepared statement of Mr. Rush follows:] Prepared statement of Hon. Bobby L. Rush Thank you, Chairman Pallone, for yielding. And thank you, Chairwoman Schakowsky, for holding this important hearing. In the 1990s, Chicago's O'Hare Airport--the world's busiest airport, at the time--became the first to install touchless faucets. This innovation's promise of making things more sanitary and wasting less water were anticlimactic, however, when it was shown that these faucets had difficulty in recognizing Black and Brown Americans' hands. While seemingly trivial, this is just one example of the real-world impediments that people of color across the country face because of the lack of representation in the technology industry. This same phenomenon has been repeated in facial recognition technology that mistakenly identified 28 Members of Congress, myself included, and in search engines that provide ads related to criminal record history when searching ``Black sounding'' names. Madame Chairwoman, it should be clear that while the issue of diversity in tech may seem like a relatively straightforward problem, its ramifications are much bigger and go much deeper than it would appear. So, I thank you for holding this hearing and look forward to hearing from the witnesses. Madame Chairwoman, I ask unanimous consent to submit a New York Times article about mistakes in facial recognition for the record. Thank you, I yield back the balance of my time. Ms. Schakowsky. And I want to thank the ranking member for her indulgence on going over some time. Next, I will yield 5 minutes to Mr. Hudson, who has been designated to take the place of our ranking member of the full committee, Mr. Walden. OPENING STATEMENT OF HON. RICHARD HUDSON, A REPRESENTATIVE IN CONGRESS FROM THE STATE OF NORTH CAROLINA Mr. Hudson. Thank you, Madam Chair. This is my first opportunity publicly to say congratulations on your chairmanship. I look forward to working with you and finding common ground where we can work together. I know you are going to provide strong leadership. Thank you for recognizing me. I would say that today we have an exceptional panel of witnesses here to examine inclusion and diversity in tech. I am proud to represent a district that has many institutions of higher education that have recognized the value a diverse workforce brings to the table. As a member of the Historically Black Colleges and Universities Caucus, I have seen firsthand how these institutions greatly contribute and prepare our students for a 21st century workforce. Fayetteville State University is a prime example of this leadership. Because of their great track record, they are a recent recipient of a $2 million grant from the project Strengthening Student Success in STEM. The project at Fayetteville State University seeks to build on previous successful efforts to increase the participation of African- American students in STEM disciplines through education and research. Additionally, we will examine ways to include other groups of individuals, including service-disabled veterans and military spouses. My district is home to the epicenter of the universe, Fort Bragg, home of the Airborne Special Operations Forces. Our community is no stranger to supporting these groups, and we understand the value that they bring to our community. Many businesses and entrepreneurs in the tech space have started in our community, including RLM Communications, a minority-owned and service-disabled, veteran-owned small business which has repeatedly been recognized for its outstanding work. The fact is that a variety of experiences and perspectives yield better results. That is exactly what diversity brings to the table. That is why I have been proud to work very closely with Chairman Rush the past two Congresses on the issue of diversity in our 21st century energy economy and preparing our students for those jobs, particularly minority students and disadvantaged groups. And I look forward to working with my colleagues on this important issue here in the tech industry as well. With that, Madam Chair, I would like to yield the balance of my time to my good friend from Montana, Mr. Gianforte. Mr. Gianforte. Thank you, Mr. Hudson, for the recognition. Getting more Americans working in tech businesses increases opportunities for all of our communities, not just in Silicon Valley. In Montana, we continually face a workforce availability challenge across most businesses. It is particularly pronounced, however, in the high-tech sector. In the software business that I built in Montana, we were always competing to get the best and the brightest. Rather than go far and abroad, however, we worked to develop people from within. Recently, I had the pleasure of spending a day with fourth- graders at Bozeman's Emily Dickinson School during their hour of code. Allowing the kids time to engage in coding early increases awareness and drew new faces into the field. Using prewritten code, these students were able to blow up their screens and make farm animals talk. They were having fun and they were learning at the same time. Another successful program for the State has been Code Montana. This class brings high school students into the computer lab. Students experiment with JavaScript and other programs to create their own apps and earn college credit. Over 90 percent of Montana parents want their kids to study computer science. Unfortunately, only 40 percent of our public schools offer this curriculum. Our company understood that we needed to grow our workforce organically, and we recognized the challenges facing our local schools. We started working with the local university, Montana State, to develop cutting-edge computer science programs. Other entrepreneurs developed classes to develop their employees for the next generation. And through organizations like the Montana High Tech Business Alliance, we created space to share ideas and address challenges facing tech in Montana. We are making progress, but there is still a lot more to do. I look forward to the testimony and the discussion as we look for solutions. Thank you, and I yield back. Ms. Schakowsky. Mr. Hudson, do you yield back? Mr. Hudson. I do. Ms. Schakowsky. OK. Thank you. So, the Chair wants to remind Members that, pursuant to committee rules, all Members' written opening statements shall be made part of the record. And now, I would like to introduce our illustrious panel and our witnesses. First, we have Mr. Mark Luckie, digital media strategist and former manager at Facebook and Twitter. Dr. Joan Ferrini-Mundy, president of the University of Maine, welcome. Ms. Jiny Kim, Vice President of Policy and Programs at Asian Americans Advancing Justice. Dr. Nicol Turner Lee, fellow at the Center for Technology Innovation, Governance Studies, at the Brookings Institution. Ms. Natalie Oliverio, CEO of Military Talent Partners. Ms. Jill Houghton, president and CEO of Disability:IN. And Mr. David Lopez, counsel at Outten & Golden, LLP, and co-dean of the Rutgers Law School. And we are missing somebody, right? Did I get everybody? OK, I guess it is Natalie Oliverio, who is not here yet. OK. So, let's begin, then, with Mr. Luckie, 5 minutes. STATEMENTS OF MARK S. LUCKIE, DIGITAL MEDIA STRATEGIST AND FORMER MANAGER, TWITTER AND FACEBOOK; JOAN FERRINI-MUNDY, PH.D., PRESIDENT, UNIVERSITY OF MAINE; JINY KIM, VICE PRESIDENT, POLICY AND PROGRAMS, ASIAN AMERICANS ADVANCING JUSTICE; NICOL TURNER LEE, PH.D., FELLOW, CENTER FOR TECHNOLOGY INNOVATION, GOVERNANCE STUDIES, BROOKINGS INSTITUTION; NATALIE OLIVERIO, FOUNDER AND CHIEF EXECUTIVE OFFICER, MILITARY TALENT PARTNERS; JILL HOUGHTON, PRESIDENT AND CHIEF EXECUTIVE OFFICER, DISABILITY:IN, AND DAVID LOPEZ, CO-DEAN, RUTGERS LAW SCHOOL- NEWARK STATEMENT OF MARK S. LUCKIE Mr. Luckie. Chairwoman Schakowsky and members of the committee, I thank you for the invitation to participate in today's hearing on inclusion in tech. From toddlers to seniors, technology has become a vital part of the lives of many Americans. However, the way we use these tech products is not singular. Gender, age, race, ethnicity, sexual orientation, disability, religion, political beliefs, geographic location, and other factors can all dramatically change the way people interact with a product. And yet, the hundreds of thousands of people who are creating these technologies remain mostly homogeneous. The tech industry continues to be populated by mostly white and Asian men. Diversity is an integral part of the fabric of America, but that is not reflected in the companies that affect the lives of millions of this country's citizens. In my time as a manager at tech companies, including Facebook, Twitter, and Reddit, I have seen firsthand the issues around diversity that continue to plague the industry. The concerns surrounding the lack of diversity at U.S. technology companies is not just about the fair treatment of their employees. What is even more alarming is the discrimination built into the products emerging from Silicon Valley companies, from the use of artificial intelligence software, more likely to flag black defendants as future criminals; Asian-Americans being charged higher prices for online test prep; apps lightening the skin of users to make them more attractive; Amazon's recruiting tool penalizing resumes that included the word ``women's,'' to Facebook's real-name policy that discriminates against Native American names and transgender people. In many of these instances, the transgressions were unintentional and later corrected, but most of these oversights can be mitigated by employing and retaining staff from diverse backgrounds in an environment that welcomes all voices. Statistically, tech companies are not doing that. In 2017, at eight of the largest tech companies in the U.S., women, on average, made up a little over 30 percent of the staff; 4.2 percent were black, and 6 percent were Latinx, according to the company's self-reported numbers. While there is a disproportionately high number of Asians who work in tech, industry data shows they are the least likely to attain a leadership role. One of the common explanations of why there is a lack of diversity in Silicon Valley is the pipeline. Tech companies argue that there are not enough women and people of color graduating with degrees in computer science. However, there are more women and people of color with tech-related degrees that are graduating than are actually being hired. The discussions around the pipeline also ignore an obvious, but overlooked fact. Most jobs at tech companies are not in engineering. At least one-third of the jobs listed in the career websites of many of the top tech companies are on nonengineering teams. There is a common refrain in Silicon Valley: ``We can't lower the bar.'' This term is widely understood to infer that black, Latinx, and women candidates are less qualified. Their hiring would be a token, putting them over more qualified white or Asian male candidates, who in some cases are actually equally or sometimes less qualified. When women and people of color are hired, they often face unwelcoming environments that upend the great work they came to do. Half of all diverse employees said they see bias as part of their day-to-day work experience, according to a recent study. Women in tech are leaving the industry at nearly double the rate as men. A person over 40 at a tech company is a rarity, and even more so if they are not in a managerial position. Despite all these challenges, we are thriving. We are leaders impacting our communities and executing the ideas that are transforming the landscape of technology and beyond. Tech companies need to recognize the greatness or risk losing some of the industry's most brilliant minds. A study by McKinsey found that ethnically diverse companies were more than 35 percent more likely to outperform their industry counterparts. And companies in the bottom quartile, both for gender and for ethnicity and race, are statistically less likely to achieve above-average financial returns. Superficial proclamations from corporate leaders are not enough. It is time to stop saying we can do better and to start being better. For Congress and this committee, more oversight of this Nation's tech companies is absolutely necessary. Continuing to learn about how the industry functions in hearings like this will lead to better economic solutions for all Americans. Tech companies must do their part, step up, and reflect the ideals of equality, democracy, and justice for all, on which this country was founded. Thank you for this opportunity to share, and I look forward to answering questions you may have. [The prepared statement of Mr. Luckie follows:] [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT] Ms. Schakowsky. Thank you very much. Next, I want to invite Dr. Joan Ferrini-Mundy for 5 minutes. STATEMENT OF JOAN FERRINI-MUNDY Dr. Ferrini-Mundy. Thank you. Good morning, Chairman Schakowsky, Ranking Member Rodgers, and members of the subcommittee. Thank you for inviting me here today. My name is Joan Ferrini-Mundy, and since July of 2018, I have been the president of the University of Maine and the University of Maine at Machias. I plan to comment on the roles of institutions of higher education in the preparation of a diverse STEM and technology workforce that is ready to solve problems and innovate through diversity, and to provide pathways, especially in STEM, for diversity. The mission of the University of Maine is to advance learning and discovery through excellence and innovation in undergraduate and graduate academic programs while addressing the complex challenges and opportunities of the 21st century through research-based knowledge. Our population on our campus of about 11,000 students in Orono, Maine, is 51 percent women. Our geographic diversity is above the national average, but we are below the national average in racial and ethnic diversity. White students comprise about 82 percent of our student body, and 12 percent of our students are black or African-American, Hispanic, or Latino, and other races and ethnicities. In addition to diversity of race, gender, and ethnicity, at U Maine we have diversity on the dimensions of age, sexual orientation, socioeconomic status, religion, disability, disciplinary background, veteran service, and experience. In Maine, 47 percent of precollege students are economically disadvantaged. Twenty-six percent of our entering students are first-generation college students. For some college students, the diverse environments of higher education are their first experience with people different from themselves and from their home communities. Our institutions need to be skilled and effective in supporting their success and opening up their pathways into the tech industries and into the STEM fields more generally. At the University of Maine, we have more than 40 different organizations and initiatives to support our diverse students. Scott Page has argued that people from different backgrounds have varying ways of looking at problems. There is certainly a lot of evidence that people's identity groups, ethnic, racial, sexual, age, matter when it comes to diversity in thinking. And as we have heard, their solutions to problems will be equitable and more effective. One function of the university is to prepare leaders and a workforce for tomorrow. And universities across the country are doing so with inclusive, cross-disciplinary STEM programs, research to better understand and ensure inclusion, support services to promote student success, curriculum and instruction designed to build from the diversity on our campuses and to engage diverse students, and through partnerships. Through the University of Maine, for example, students, faculty, and staff from around the world and from very diverse backgrounds come together in our Graduate School for the Biomedical Sciences and Engineering to participate in convergence research. They work in molecular and cellular biology, bioinformatics and genomics, toxicology, neuroscience, and biomedical engineering, to address some of tomorrow's biggest challenges. And the diversity that they bring to those challenges is absolutely essential. Our campus Center on Aging promotes and facilitates activities on aging and aging-related education and training programs. Our Center for Community Inclusion and Disability Studies partners with the community to enhance the lives of individuals with disabilities and their families. In Maine, within 1 year of high school graduation, 50 percent of economically disadvantaged youth are enrolled in higher education compared to 75 percent of noneconomically disadvantaged youth. Enabling all students to be able to attain degrees is critical for the diversity that we discuss here today. The University of Maine is addressing this through a new First Year Student Initiative. Some of our focus is directly in the curriculum. With funding from the National Science Foundation INCLUDES Program, U Maine anthropologist Dr. Darren Ranco and colleagues are developing the Wabanaki Youth in Science Program to bridge inclusion in postsecondary education through the sciences. The team is developing a course that brings together traditional ecological knowledge and Western science for undergraduates. The success of this ongoing effort will depend upon diversity and student engagement with ideas that are unfamiliar and with experts who bring diverse experiences. I would like to also mention that the University of Maine system partners with Educate Maine in a project to match students with internships and full-time jobs in computing occupations. And we hope to bring underrepresented minorities, women, and rural students to pursue a variety of experiential learning opportunities. In closing, our Nation's diversity is a resource for learning and for solving the most complex problems of our times. As technologies and advances lead to changes in our workplace, new generations of STEM professionals will bring their collective diversity to bear on improvements and innovations. U.S. higher education has a key role to play in ensuring the pathways for inclusive learning environments, so that those professionals are ready to make a difference. Thank you for the opportunity to speak with you today. [The prepared statement of Dr. Ferrini-Mundy follows:] [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT] Ms. Schakowsky. Right on the dot. Thank you. Ms. Jiny Kim, you are now recognized for 5 minutes. STATEMENT OF JINY KIM Ms. Kim. Good morning, Chairwoman Schakowsky, Ranking Member Rodgers, and members of the subcommittee. Thank you for the opportunity to testify on the importance of diversity in the technology sector. My name is Jiny Kim, and I am vice president for policy and programs at Asian Americans Advancing Justice, AAJC, a national civil rights organization dedicated to advancing the civil and human rights of Asian-Americans and building and promoting a fair and equitable society for all. The economic rationale for diversity has been well- documented in numerous studies. Companies in the top quartile in terms of racial diversity are 35 percent more likely to have financial returns higher than the national median in their industry. But, despite this economic reality, women and people of color have historically been excluded from both the rank and file and from positions of leadership within tech companies. According to the EEOC, African Americans and Latinos were underrepresented in the tech sector by 16 to 18 percentage points compared with their presence in the American labor force overall. While there is a higher representation of Asians in the tech workforce, they are still underrepresented in nontechnical roles compared to their presence in technical roles, and they are disproportionately left out of C-suite positions. While the effort companies are making to provide transparency in their diversity data should be appreciated, there remain issues in how that data is reported. Asian- Americans and Pacific Islanders include over 50 different ethnic groups and over 100 languages and dialects. Yet, companies fail to disaggregate the data, resulting in overlooking the most underrepresented and historically marginalized AAPI communities with lower levels of educational attainment, higher rates of poverty, and larger populations with limited English proficiency. When these groups are left out, those efforts by industry and other stakeholders to encourage recruitment and build pipelines from diverse communities remain incomplete. Not surprisingly, tech companies have developed digital tools to review the myriad applications for positions in their companies. The problem with this approach is that the ideal profile being used as a model reflects a majority white culture and the resulting unconscious bias. Issues are not limited to recruitment, and greater effort is also needed to retain employees of color and women. Some tech companies have taken the important step of reporting attrition rates of employees from diverse backgrounds, as well as supporting their employees through mentorship programs and employee resource groups. We applaud these efforts as positive steps toward understanding what is needed to retain diverse staff and eventually place them in the leadership pipeline. Now it is a common understanding among civil society organizations that the prejudice, ignorance, and the hate we combat in real life live in the digital space as well. Tech companies that foster a majority white male employee base feed their own biases into the machines they create. In the criminal justice system, we see disturbing examples of algorithmic bias. Courts have begun using predictive software to sentence convicted individuals. ProPublica published an account of two individuals who separately committed shoplifting. One was African-American, and the other was white. When a sentencing algorithm was used to predict the likelihood of each committing a future crime, the African- American individual was rated a higher risk, even though she had only committed misdemeanors as a juvenile, while the white individual had previously been convicted of attempted armed robbery. Two years later, the computer algorithm was proven wrong, with only the white individual having committed a felony. Further alarming is facial recognition technology. In 2015, this technology came under scrutiny when software incorrectly categorized photos of African Americans as primates. Despite this incident, companies have still failed to take adequate action. Studies published as recently as last year found that facial recognition algorithms had significantly higher error rates detecting the gender of darker-skinned individuals compared to lighter-skinned individuals. Given the magnitude of the impact of lack of diversity in tech, a serious culture shift must take place. And civil rights organizations like Advancing Justice, AAJC, have already begun to play their part. Last month, we joined more than 40 advocacy groups in sending a letter to congressional leaders urging them to put civil and human rights at the center of the digital privacy discourse. And tech companies have begun engaging our organizations on diversity and inclusion issues and taking part in civil rights audits. The tech sector has transformed the way we communicate and connect with one another. We must ensure that the development of technological products, services, and experiences leave no one behind and do not harm communities of color. In order to do so, employees who create these innovative tools must reflect the diversity of the communities that the companies seek to reach. Thank you for this opportunity. [The prepared statement of Ms. Kim follows:]\1\ --------------------------------------------------------------------------- \1\ A February 2017 report entitled ``Breaking the Mold: Investing in Racial Diversity in Tech'' submitted by Ms. Kim has been retained in committee files and also is available as part of her written testimony at https://docs.house.gov/meetings/IF/IF17/20190306/108901/HHRG-116- IF17-Wstate-KimJ-20190306.pdf. [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT] Ms. Schakowsky. Thank you. We do have that letter in the record from last year, from the last hearing rather. So, thank you for that. And next, I want to recognize Dr. Nicol Turner Lee for 5 minutes. STATEMENT OF NICOL TURNER LEE Dr. Turner Lee. Thank you. Good morning, Chairwoman Schakowsky, Ranking Member Rodgers, and members of the subcommittee. I am encouraged by your interest in this topic, particularly as some of the members of the subcommittee have worked to diversity your own staff. Let me start just by stating again what Congressman Lujan summarized from my written testimony. The absence of diversity among the people who make decisions around products and services, along with the markets that these companies and the tech serve, will ultimately lead the U.S. to abysmal failure. With the U.S. population predicted to become minority white in 2045, tech companies that do not fully embrace diversity will ultimately compromise the quality of future technologies and make it difficult for all people to gain the benefits of the digital revolution. Today, existing and emerging technologies are helping to solve complex social problems through automation, advanced scientific research, and artificial intelligence, while disrupting legacy industries and widely accepted norms. Yet, despite this growth, African Americans and Hispanics remain vastly underrepresented in the computer and mathematical fields, 7.9 percent and 6.8 percent, respectively, compared to whites, which are over at least 12 to 14 percent. In fact, African Americans and Hispanics are the most underrepresented in certain tech jobs by nearly 50 percent. Less than 5 percent of the tech workforce in social media companies is African-American, with similar findings for Hispanics and certain Asian-American populations. The irony here is that 35 percent of Hispanics and 24 percent of African Americans have no other online connection, except through their smartphones and mobile devices, compared to 14 percent whites, but they are connected to the applications and the platforms that actually run off these devices. Without them, they have no other way to live, learn, earn, vote, and network. These mismatched realities also make these populations most susceptible to digital disruption when the jobs that they once held are automated and eliminated or predatory products and services are marketed to them online on an ongoing basis. So, this is why diversity matters. In certain sectors of tech, there is a talent pipeline problem, leaving empty pockets of workforce diversity in board, C-suite, and other leadership positions. And this human resource problem ultimately impacts the design, implementation, and evaluation of products and services, some of which are collectively profiling, surveilling, and even discriminating against protected classes. That is why we are here today. I am going to just summarize in my closing remarks three things that I have actually put forth that I think Congress and the tech sector should do. First, tech companies must be more deliberate and systematic in the recruitment, hiring, and retention of diverse talent and change the sources for where they find talent of color. Historically Black Colleges and Universities and Hispanic-serving institutions are often discounted in remedying pipeline concerns. Yet, 22 percent of African-American college students graduate from one of the 101 HBCUs and offer some of the Nation's most gifted talent which are prepared to work in Silicon Valley. Yet, those colleges are not the source for where we actually look for talent. In addition to that, they may be challenged by the resources that HBCUs and HSIs find. The appropriations that go to those universities and colleges are often comparable to others; therefore, reducing their ability to entice a tech center to actually hire them. Congress, we must do better than that. Second, tech companies must explore ethical and collaborative frameworks that explore the intended and unintended biases of algorithms and deploy solutions that quell these biases. With big data being collected in real time from users at all times, people are now being denied credit based on their web-browsing history or aggregated predictive analytics are wrongly determining a person's suitability for employment or applying a longer prison sentence. These are deplorable, and we need to work together, as my colleagues have said, to increase the pipeline, so we can make less of these mistakes. Even among members of the Congressional Black Caucus, facial recognition technology wrongly associated them with arrest records 90 percent of the time, and I know my distinguished members of the CBC aren't those people. And finally, I would say this inattentional blindness is a problem in the tech sector that should no longer be tolerable. The strength of the online economy proves that it is no longer insulated from the guardrails designed for other regulated industries, especially those that establish baseline protections against discrimination. Congress should consider review and the potential modernization of civil rights law and apply them to certain online cases. We did it in the case of housing. We did it in the case of civil rights. We have done it in the case of equal opportunity, and we should do it in the online space. Thank you very much, and I look forward to your questions. [The prepared statement of Dr. Turner Lee follows:] [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT] Ms. Schakowsky. Thank you very much. Next, we will recognize Ms. Natalie Oliverio. You are recognized for 5 minutes. Welcome. STATEMENT OF NATALIE OLIVERIO Ms. Oliverio. Thank you so much for allowing me to testify before you this morning. My name is Natalie Oliverio, and I am a post-9/11 Navy veteran, the founder and CEO of Military Talent Partners. And too frequently, too often, veterans and military spouses are left out of the diversity conversation. It doesn't naturally occur to you to think of them as diverse individuals, but less than 1 percent of our country today serves. So, when you serve in the military or you are a spouse supporting your significant other as they serve, time doesn't stand still. The world keeps moving without you while you are serving your country. The military is its own business and it has ever job type, skill, and capability to run on its own, and those skills really prepare veterans to do anything. But I talk with companies every single day who feel that they don't have jobs suitable for veterans. They don't have security-type or protective work, which is what they believe that veterans are qualified and capable of doing. There have been massive steps forward in the programs offered, the trainings available, and the abilities that veterans can then leverage to their second careers. But not everyone knows about them because they are just left out of the conversation, even more so for military spouses. For veterans, unemployment isn't really the issue anymore. It is more of underemployment. Meaningful careers make all the difference. From an extreme case of contributing to 22 suicides a day in the veteran communities, meaningful careers can put a stop to that. Meaningful careers for military spouses help them provide for their families when today's economy really needs a two-income household. Putting your life on hold to support your spouse as they are service member is a major sacrifice, but in today's job market they are seen as risky hires or job-hoppers, but that is not the case at all. So, we have a lot of stereotypes to overcome, but those challenges can be overcome by adding veterans and military spouses to the conversations around diversity and inclusion. I, myself, had a very difficult transition from active duty post-9/11, and I thought I had it all figured out, but I didn't know what kind of resources were available to me and what kind of possibilities existed. So, I struggled for years to find my niche, and that has been my motivation to lead the way in mentorship and coaching for all active-duty transitioning service members, military spouses, and Gold Star families, to help them find and obtain the meaningful careers for them. But that is just one small step. There is a lot of work to be done. It is really up to Congress to lead the way for more veteran initiatives, and not just the initiatives and programs that are benefitting us today, but to make it known, widespread, so there is no question about the capabilities that a military spouse or a veteran brings to the corporate world today. Thank you so much. [The prepared statement of Ms. Oliverio follows:] [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT] Ms. Schakowsky. Thank you. And now, Ms. Jill Houghton is recognized for 5 minutes. Thank you. STATEMENT OF JILL HOUGHTON Ms. Houghton. Chairman Schakowsky, Ranking Member McMorris Rodgers, members of the committee, thank you for the opportunity to testify today. My name is Jill Houghton. I am the president and chief executive officer of a nonprofit called Disability:IN, and we exist to help business achieve disability inclusion and equality. We have over 50 affiliates across the country. We represent more than 170 major Fortune 1000 corporations, and almost one-quarter of those corporations operate within the technology industry. My testimony is rooted in my personal experience. I am a female leader and I have a nonapparent disability. And I really want to focus on three issues today. I think it is really important that you remember that disability is an important component of diversity. No. 2, disability drives business performance. And No. 3, disability inclusion drives innovation. Disability knows no stranger. Race, gender, ethnicity, sexual orientation, gender identity, it can happen to any one of us at anytime in our life. According to the Centers for Disease Control and Prevention, 1 in 4 Americans live with a disability, and disability is often forgotten. We are said to be the silent ``D'' in diversity. And yet, there are 61 million Americans living with disabilities. So, we are an integral part of diversity. When the Americans with Disabilities Act was passed in 1990, that was opportunity, economic opportunity, for people with disabilities, but the one thing that it couldn't do was legislate attitudes. And so, we know that, when we look at the Bureau of Labor Statistics in January of 2019, that the labor force participation rate for people with disabilities was 20.5 percent versus people without disabilities of 68.3 percent. At Disability:IN, we have committed our self to helping corporate America with data and insight. In that spirit, we have joined forces with the American Association of People with Disabilities to create something called the Disability Equality Index. This is the Nation's trusted disability inclusion benchmark, because business wants to do better. and it looks at things like leadership and culture, employment practices, community engagement, enterprisewide access, and supplier diversity. The technology industry actively participates in the DEI, and we are growing by 30 percent every year. The companies that score an 80 or above are publicly acknowledged on our website at disabilityequalityindex.org and ranked as the best place to work for people with disabilities. But I will tell you, even the companies that are scoring a hundred would be the first to tell you that they don't have it all figured out, that they want to do better, but there is a lot more work to do. Using the Disability Equality Index, we teamed with Accenture, because what we know, if we are going to take disability inclusion and diversity to the board room, to the C- suite, we need the business case. So, we teamed with Accenture. They studied the first 4 years of data. They worked with Vanguard and algorithms. What they found in getting to equal the disability inclusion advantage is that, on average, companies, leading companies, that are driving disability inclusion rated 28 percent in higher revenue, double the net income, and 30 percent higher economic profit margins than their peers. Disability inclusion impacts business performance. With that concrete evidence, we confirmed that disability inclusion is good for business and investors are viewing it as the next frontier in environment, social, and governance investing. We rolled out that report on the floor of the Stock Exchange and Wall Street. And the Comptroller DiNapoli from New York State has issued a letter--he runs the third largest pension fund in the Nation--to the top Nasdaq companies calling on them to ask what they are doing around disability inclusion. And last, disability inclusion drives innovation. We like to say that inclusion and disability, at the nexus is accessibility, and without accessibility, we have got nothing. And so, that is very important, that the tech sector continue to focus on building accessibility and driving disability inclusion. Thank you. [The prepared statement of Ms. Houghton follows:]\2\ --------------------------------------------------------------------------- \2\ A report entitled ``The 2018 Disability Equality Index: A Record Year for Corporate Disability Inclusion and Leadership'' submitted by Ms. Houghton has been retained in committee files and also is available as part of her written testimony at https:// docs.house.gov/meetings/IF/IF17/20190306/108901/HHRG-116-IF17-Wstate- HoughtonJ-20190306.pdf. [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT] Ms. Schakowsky. And next, Dean David Lopez, you are recognized for 5 minutes. STATEMENT OF DAVID LOPEZ Mr. Lopez. Thank you very much, Chairwoman Schakowsky, Ranking Member Rodgers, members of the subcommittee, for inviting me to this very important hearing. My name is David Lopez, and I am currently the co-dean of Rutgers Law School in Newark. Over the last 100 years, Rutgers Law School has stood as an exemplary model of a public institution that both welcomes and promotes diversity, meaningful sociability, and leverages the law to achieve equality of opportunity in the public, private, and nonprofit sectors. We believe we have normalized the idea of opportunity. From 2010 to 2016, I was the longest-serving general counsel of the U.S. Equal Employment Opportunity Commission, twice nominated by President Barack Obama and twice confirmed by the U.S. Senate. I was the first Latino selected to this position, and in the capacity as general counsel, I led the litigation program charged with enforcing Federal antidiscrimination statutes nationwide. So, I come here both as an educator and as a lifelong civil rights practitioner. During the last 30 years, technology has transformed our economy and changed our daily lives--how we work, how we learn, how we make decisions, how we play. The tech industry has produced remarkable tools and resources, providing us with social media and new ways to connect with others, as well as instant access to huge amounts of information. But, as we have heard, it is no secret that the tech industry has suffered with the persistent problem of the absence of diversity. We have heard many of the Members here today, from many of you, from many of the panelists. And this all comes at a time when tech jobs are growing rapidly in our economy. To use the words of a very common phrase today, ``The tech industry operates in a bubble.'' It operates in a bubble. It operates without the cross-currents of thinking, from reflecting the beauty of this country. Of course, ideals of living in a discrimination-free society with equal opportunity, these are bedrock principles central to social mobility and the American dream. The elimination of arbitrary barriers ensures that hard work matters, that investing in your dreams matters. It is well established that racial, gender, and other types of diversity in the workplace has a positive influence on teams, and we have heard many of those studies. Diverse teams are more productive. Teams that are made up of individuals of diverse backgrounds are more innovative, generally make more error-free decisions. Further, there is convincing evidence that increased diversity in the workplace leads to higher revenues and increases innovation. One of the problems I believe that the tech industry faces is the problem of implicit bias. The science of implicit bias is recognized as the automatic associations of stereotypes or attitudes about a particular group. One study demonstrated implicit bias by showing that resumes with more white-sounding names received requests for interviews 50 percent more frequently than the same resume with more African-American- sounding names, but with equal or better qualifications. A number of recent studies also suggest that isolation and bias influenced women leaving STEM careers. Often, it is not simply the choices that employees make that influence the careers, but the workplace environment that drives denial of opportunity. Now let's talk a little bit about the products. And we have heard many of the stories here today about the end results of perhaps the absence of diversity. Big data analytics allows your employer to know whether you are pregnant even before you disclose it. That is against the law. In one high-profile incident, one retailer, drawing on consumer data, knew a young woman was pregnant before her parents did. Software used by many police departments across the country that determines the likelihood of recidivism has been shown, as discussed today, to have a bias against African Americans. Companies are using algorithms to determine who is likely to default on a loan or recommit crimes, despite the algorithm's tendency to reflect society's bias towards racial minorities. So, all of this demonstrates that this is a very important discussion. Diversity is important not only because it is the right thing to do and it is a reflection of the American dream; diversity also has consequences on these tools and these products of predictive analytics and artificial intelligence rapidly changing our environment, and the way that we live and that we play in our society. The public university has a very special role in addressing these issues because a public university provides an opportunity for interdisciplinary learning, for computer science, that enshrines values of equal opportunity, fairness, competition, and justice. And I welcome your questions. Thank you. [The prepared statement of Mr. Lopez follows:] [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT] Ms. Schakowsky. Thank you so much. This has been really a wonderful panel that I think underscores how diversity is not just a side issue. And now, we will move to a number of questions. Each Member will have 5 minutes to ask questions of our witnesses, and I am going to start by recognizing myself for 5 minutes. I just want the panel to know that there are a lot of things going on today. The fact that there are a number of empty chairs is not indicative of lack of interest in what you are saying. All of your statements will be in the record, and I expect people will be coming in and out. So, I am going to focus on women. Just 31 percent of the employees at Facebook are women, and that's also true at Apple. The number of women who work in technical roles at these companies is even lower. We also see that the female share of computer science degrees has actually dropped from 28 percent to 18 percent between 1993 and 2016, while at the same time the tech industry is booming and continuing to grow at an unprecedented rate. This isn't a problem that will be solved overnight. But, starting at the top, getting more women into visible technical positions, providing role models for young women, college students deciding on choosing a technical path of study, seems to be a very good strategy. So, Dean Lopez, intentional efforts to reduce the bias that is not necessarily deliberate in recruiting have dramatically increased opportunities for women in other fields. For example, orchestra auditions where the musician's gender was hidden has increased the hiring of women by over 25 percent. What are some ways that tech companies can change their recruitment techniques in order to adapt to the need of diversity in the workforce? Mr. Lopez. I thin the most important thing a tech company can do really is provide meaningful leadership at the top. If tech companies want to be diverse, they need to have the leadership making very clear pronouncements that they will examine every aspect of the work culture, the evaluation system, the recruitment system, the promotion system, to make sure that it is free from gender bias. And that means looking at issues of implicit bias. You see this particularly in pay disparities based on gender, where you often have systems that are almost systematically stacked against women. So, it really takes a clear statement from leadership that diversity is important, that inclusiveness is important. The other thing that is really important is to really examine how you conduct recruitment. Recruitment is often based on sort of the tap on the shoulder, the old boys' club. It is really important, I think, for high-tech companies to examine their networks. As the co-dean at a public university, as a graduate of Arizona State University, I certainly understand the grit and the talent that you get out of those universities, but often the Silicon Valley recruits in a very, very narrow way, right? And then, the last thing I want to mention is that, as diversity relates to the end product, we have seen Amazon, for instance, they had to stop a hiring tool because they found out that the hiring tool itself was biased against women. The algorithms they used were biased against women. We have seen lawsuits recently involving Facebook's dropdown box where women were excluded from certain job advertisements in traditionally male industries. So, there is a connection between the two. Ms. Schakowsky. Thank you. Funding for female entrepreneurs is minuscule compared to men. Women receive just 2.2 percent of the venture capital investment. This is despite the fact that women-founded businesses generate more than two times the revenue per investment dollar than businesses founded by men. Ms. Kim, what do you think is happening here? What is going on? Ms. Kim. Much like the other economic reasons for having diversity, you are seeing something that doesn't make sense. It is the system that exists that needs a complete culture shift in how to--I mean, whether it is employee recruitment or whether it is investing in programs to recruit from diverse populations, as well as investing in building those opportunities for entrepreneurs as well. It doesn't make sense, what is going on, because there is a clear economic rationale to hire and promote from communities of color, hire and promote women, and invest in businesses led by people of color and women. And so, we call on the tech sector to examine what within the culture is going on, and, also, to engage with civil society organizations, and civil rights organizations, in particular, to talk to us about what is going on in the hiring and investment practices that results in these very harmful impacts. Ms. Schakowsky. Yes, I wish I could get to--5 minutes goes really fast. Maybe at the end I can open it up to others on the panel. But, in the meantime, I want to yield for questions to our ranking member, Ms. McMorris Rodgers. Mrs. Rodgers. Thank you. Thank you, Madam Chair. And thank you, everyone, for being here. Excellent testimony today. I wanted to start with Ms. Houghton. I wanted to ask you to speak some more to how companies can build an inclusive workforce for people with disabilities, and address both the physical and intellectual disabilities, and what the benefits are that you see as a result. I also wanted to ask you to speak and share some of the examples of the impact of having people with disabilities in the workforce and how that ensures products and services are accessible to everyone. Ms. Houghton. Thank you, Congresswoman McMorris Rodgers. I think that what we are seeing is really driven in tech. It is companies like Microsoft and SAP and DXC Technology that have created these inclusive hiring programs that are based on the premise that there is bias in their hiring process, and they want to tear down those walls. And so, they have created these hiring programs where they are sourcing talent with disabilities, with all different kinds of disabilities, and bringing them in, perhaps rather than in a typical interview, coming in and maybe doing a two-week program. And maybe the interview is with LEGOs and like a different kind of a process. Because what they recognize is that the traditional models are screening the talent out. And so, they have grown these inclusive programs, these Autism at Work programs, and they are putting everything out in the public domain to try to help their peers. I think what they are experiencing, as a result of this, is that this talent is coming in and helping them develop new products and tools and make things more accessible. Filing for patents, things that wouldn't have happened if they hadn't torn down those walls. Mrs. Rodgers. Would you speak briefly as to the impact of job coaches or accommodations, the internships, just very briefly? Ms. Houghton. Yes, people with disabilities, we come in different shapes and sizes. The on-the-job supports when individuals have the opportunity to perform with the right support, they far exceed their peers. Their productivity, their decrease in absenteeism, they stay. Mrs. Rodgers. That is great. Thank you. Thank you for being here. Dr. Ferrini-Mundy, I wanted just to ask you to speak a little bit more about how you see the commitment to science, technology, engineering, and math helping create the pipeline, and are you seeing that translate into the opportunities with the tech companies, in particular? Dr. Ferrini-Mundy. Thank you. Absolutely. The STEM field, science, technology, engineering, and mathematics, writ large, are often a key foundation for people who will pursue careers in the technology areas. And so, making certain that that basic preparation--and I will speak specifically about mathematics; it is my own field. And also, it is an underpinning for so much of what goes on in the tech industry. Making certain that our approach to engaging people in mathematics, to instruction, is inclusive, that it attends to differences and draws on those, and supports students to be successful, to draw on their grit, because mathematics is not necessarily seen as an easy kind of pursuit. To continue, I wanted to just tack on a little bit on these questions about internships and how companies can be more fully engaged with diversity. Universities can be wonderful partners and are across our Nation, our public universities, in particular, with the private sector. And we have found that U Maine, for example, in our engineering college, that a large percentage, maybe 75 percent of our students have actual internships in a variety of high-tech sectors, and those internships turn into positions. And so, getting to know students and making opportunities for diverse students early on is really crucial to this whole business. Mrs. Rodgers. Thank you. And finally, Ms. Oliverio, would you speak, just in the 20 seconds I have left, just what do you see veterans--what are the unique qualities that veterans can add to this conversation? Ms. Oliverio. Veterans are so unique. They are natural leaders and that leadership is cultivated in an authentic way throughout their military service. They are resilient. They can adapt and overcome to any situation and any challenge. If tech can just meet them where they are, they will be able to diversify on their own. Mrs. Rodgers. Great. Thank you all. I yield back. Ms. Schakowsky. I recognize Representative Castor for 5 minutes. Ms. Castor. Thank you, Chairwoman Schakowsky. Sharing diversity and inclusion in the technology sector workforce and products is critical to a thriving industry. Unfortunately, the U.S. Congress and the industry itself have overlooked this topic for too long. So, I am glad we are holding this hearing today. Thank you very much, and I want to thank the witnesses for your expert advice on what Congress can do to make the workers in the tech sector more representative of America. Because I believe, when that happens, business will thrive and consumers will benefit. In addition to your testimony, I have seen report after report that has been rather troubling about technology adversely affecting communities of color. Companies have given different prices and credit card deals to consumers based upon location, which can mean white neighborhoods are offered better deals and prices than minority neighborhoods. Supposedly unbiased algorithms that companies use for a variety of different processes have been shown to produce discriminatory results. Facial recognition software often cannot accurately recognize people of color. Ms. Turner Lee, these are just a few examples, but they illustrate a larger pattern in tech of discriminatory products and processes. What are some of the specific policies the tech industry could adopt right now to fix this, and what should Congress be doing? Dr. Turner Lee. Yes, thank you for just acknowledging all those examples. Because as we go deeper and deeper into the tech space, I think we are going to see more of these. I think it is important, as I recommended and, again, put into my written testimony, that we think about a guardrail that we have now, which are the antidiscrimination laws. I think as we see more of these offers become discriminatory or produce a discriminatory output, people being denied credit because of the fact that their web-browsing history suggests that they are not creditworthy, higher education using algorithmic decisionmaking on whether or not kids should be accepted into college, those types of things have, I think, implications that we have not really looked at in connecting the physical and the digital spaces. So, I would implore Congress to just have a review and analysis of what those nondiscrimination laws are and see if there is any connection to what we actually see in the digital space that can generate these unintended consequences. I think that is the first. I would also just add real quickly, I think there will be innocuous cases, as it was suggested, where the training data may not be correct and companies themselves will self-regulate. I think those conversations still need to be had. Google voluntarily removed payday ads from their search query, just to make sure that low-income people were not being dragged into this pathway of inequality. I think we need to see more of that, and I think Congress can actually use the bully pulpit in some ways to suggest that those conversations should happen, as well as collaborations with civil society, who actually see the outputs of this. It is the technologist that oftentimes sits within the vacuum, and civil society groups like AAJC, then, sort of have to clean it up. And then, Congress has to somehow get in the middle of these conversations. I think more collaborative dialog to best understand how these ecosystems work and the application of guardrails that we have in our favor can actually help quell some of these biases. Ms. Castor. Thank you very much. Mr. Luckie, isn't there also a role for people of diverse backgrounds in decisionmaking positions in these tech companies? And how do we encourage that and what are the barriers that prevent that from happening now? Mr. Luckie. I think there is a lot of focus on the managerial positions and having someone at the top that will filter down and make a workplace more diverse. I think it is more important on the employee level to have multiple people in the room who can say, hey, are we testing on this particular audience; have we thought about this particular impact on this community? As I said in my opening statement, making sure that those voices are being heard, that there is an equal opportunity for people to share those concepts, and it is important for tech companies to do an audit of these individual teams and understand where are the gaps in diversity, not just in the company overall, but on the individual teams that are all impacting the company's overall goals. Ms. Castor. Thank you very much. I have a few other questions. I had a group of neighbors from Florida come and visit me who happened to be blind, and they had a number of suggestions. My time is running out. So, I am going to submit those to you for the record and ask you to please send in your specific answer to those. Thank you. And I yield back. Ms. Schakowsky. Thank you. And now, 5 minutes to Mr. Guthrie. Mr. Guthrie. Thank you, Madam Chair. I appreciate it very much. Thanks for this hearing. This is a very important hearing. A couple of things. One, and this is for Dr. Ferrini-Mundy, or anybody that would like to answer the questions, but I will focus on you. I understand that Facebook, Google, and all the tech companies have people in marketing in all different degrees, but I am going to focus on the STEM side of their businesses. I think the Chair just said in her questions that, as we need more computer science people, as that seems to be the initial higher, where big money is, people in tech fields, if I heard you correctly--I know you said it correctly--but if I heard you correctly, 28 percent of the computer science used to be women and now it is down to 18 percent. So, I guess my question is, does the tech workforce in Silicon Valley or in tech, the tech people, does it reflect the people in the tech programs, the demographics? So, is it kind of they are hiring who we are training or educating as a nation? And if so, how do we get more people into it? How do we get a more diverse STEM populace, so that there will be a more diverse technical? Because the other side is, if it doesn't reflect that, that means they are just really being biased in who they hire. Of course, I am talking on the STEM side of their business, their employees. Dr. Ferrini-Mundy. Right. Mr. Guthrie. OK. Dr. Ferrini-Mundy. Thanks. And so, there are several questions in there. I think in terms of the tech workforce, others in this panel are more expert in the dispersed expertise across that tech force, because I suspect it comes from a variety of areas. That all said, we must do better in higher education to attract people to these STEM fields, to computer science, and to make their time in higher education much more inclusive, so that they are a part of the groups that are, then, going to be taking on these product questions when they get into the tech workforce. Mr. Guthrie. How do we get more people, a diverse group, into the---- Dr. Ferrini-Mundy. Right. Mr. Guthrie [continuing]. Because what happens in tech school, not just computer, if you are talking about a manufacturing company in Kentucky needs a computer science, I mean a numerical control person, whether it is male, female, or whatever, they can't find them, and we have all these people not going to secondary schools. So, there seems to be a breakdown in the market---- Dr. Ferrini-Mundy. Right. Mr. Guthrie [continuing]. For getting people into the right--no matter who they are. Dr. Ferrini-Mundy. At U Maine, I just learned that about 47 percent of our students are in STEM fields, which is high. What that means is that we are creating pathways from our secondary schools that are welcoming. We do an early college program that gives students the opportunity to study with our faculty before they get out of high school. I think a lot of it is about pathways and helping students be able to see themselves in these careers, see them as meaningful career options. And a lot of that has to do, then, with real-world kinds of problem-solving as undergraduates, so that it is real-world learning outside the classroom in internships, in clinical experiences, that put them together with people in these fields. A lot of collaboration with K-12 is crucial for this, too. It begins very early. So, all of the work in coding, for example, that we heard about is one piece of, I think, an important systemic effort to get people interested, to get diverse people interested in STEM. Mr. Guthrie. I have got a few seconds. I want to ask another question. Anybody want to comment on how we get more people into, a more diverse, educated group to come out, so a pool? Dr. Turner Lee. Yes. No, Congressman, I think it is a great question. I mean, overall, we have a national shortage of tech workers. So, let's just start there. I mean, in programming, we have seen that in government where years ago there were 10 to 15 thousand people we couldn't employ in cybersecurity tech jobs. So, if we look at the national shortage of where we are as a country, and then, you trickle that down to diversity, it becomes even more problematic, right? But I think what we are seeing is this movement in colleges and universities to sort of focus on computer science, which I think may become a better shift. I mean, I have seen Members of Congress, your colleagues, sort of introduce computer science as a national initiative. It has not been a national priority. People, you know, they change where they are in terms of their leaning towards STEM. There are studies that say, with African-American children, if a young African-American boy is not actually focused on math by sixth grade, it is less likely that he will pursue a STEM career when he goes to college. That is why I say I think it is important for us to look at the sources of where we are recruiting students and build up where there is a possibility of more appropriations in these programs, opportunities. It is also important---- Mr. Guthrie. I do have one more quick question I want to get to. Dr. Ferrini-Mundy. OK. Mr. Guthrie. I apologize for that. Ms. Oliverio, I wanted to ask you a question about veterans. I served in the military myself. But I think you said to maybe Congresswoman McMorris Rodgers, you said, ``Tech will meet veterans where they are and they will diversify themselves.'' I mean, what does that mean, tech needs to meet veterans where they are? Ms. Oliverio. By more fellowships, apprenticeships, and opportunities for veterans to bridge their skills gap in a field such as coding. Coding is wildly popular, and there is a lot of professional opportunity across corporate America to get a job in coding. But that is one major skill that is not utilized in any branch of service. Mr. Guthrie. And 5 minutes does pass too fast. [Laughter.] So, thank you, Madam Chair. Thank you for your answers. Ms. Schakowsky. Now I recognize Congresswoman Kelly for 5 minutes. Ms. Kelly. Thank you, Madam Chair. Algorithms are the undercurrent of the internet. So much of what we do online is run by automated machine-learning algorithms. But it has become clear, as we have talked about, that bias of all kinds permeates many of these algorithms. In his written testimony, Mr. Luckie pointed to several examples of bias output of algorithms. What makes this even more concerning is that, apparently, no one, often not even in the engineers and computer scientists creating the algorithms, really know how these machine-learning algorithms work. All they really know is what datasets are used to train the algorithms and what results come out of the other end. As we often hear, garbage in, garbage out. Ms. Kim, can you expand on how the decisions made by algorithms can hurt vulnerable communities? Ms. Kim. Thank you, Congresswoman. Ms. Kelly. You're welcome. Ms. Kim. In terms of, again, the specifics of the technical aspects of why things happen, it is upon industry to let us know, have more transparency, and work and engage with us in terms of civil rights organizations and communities of color to let us know why these things are happening. But, in terms of the examples that have happened, we have seen, for instance, in Oakland, California, the police department using predictive software to send police to neighbors that are more often than not communities of color, regardless of the actual crime rate of those neighborhoods. You see examples again and again like this. And it is upon us, it is our job as civil society organizations to raise these issues to tech companies, but the tech companies must engage, and many have. And we appreciate companies that have engaged in civil rights audits and other opportunities to raise these concerns, and often to raise concerns before they become actual problems. And so, we look forward to additional engagement. Ms. Kelly. So you feel the companies need to be more proactive? Ms. Kim. Absolutely. Ms. Kelly. Mr. Luckie, you said that some of the more major incidents that you listed can result in bad publicity, which can alienate customers, leading to profit loss. Is it fair to say that there are biased outcomes that have not been identified? Mr. Luckie. Absolutely. One of the things about working at Facebook, in particular, is that you don't see the fires that the company has put out before it gets to the public. And there are whole teams that are just working on getting those out of the public eye. What I will say is that companies like Facebook think about the best possible uses of their platform and not the worst. That is where you see issues like Russia and hackings and privacy, and then, they become issues that they have to fix later down the road. Ms. Kelly. So, are you saying that companies, again, need to be more proactive and not just reactive-- Mr. Luckie. Absolutely. Absolutely. Ms. Kelly [continuing]. When something happens or when the press is looking? Mr. Luckie. And having more people in the room from different backgrounds will aid in that. Ms. Kelly. OK. Targeted online advertising has become so sophisticated that advertisers can skirt Federal law by using interests as a proxy for disability, race, or other protected traits. Mr. Luckie, what do platforms need to do to address these loopholes and fight less blatant forms of discrimination? Mr. Luckie. It is really about the education and making sure that that is being disseminated from the top. Too often what is happening is this is happening on a ground level, where employees are fighting the good fight and educating up, rather than that happening top-down. And so, it has to be a priority for leadership and them to be vocal and, also, to hold people accountable in order to make sure that these things aren't happening. Ms. Kelly. Dr. Turner Lee, do you have anything more to add? Dr. Turner Lee. Yes, I would add, the interesting area that we are in, Congresswoman, right now is the fact that we do not have demographic data collected about us by technology companies. And so, what you are suggesting is that use proxies, your zip code, maybe your profile picture, things that actually are subjective measures to sort of come at your identity. Because a lot of what we see in algorithms are inferential circumstances, right, it is inferring from my purchasing behavior where I visit, who I speak to and connect to, the type of person that I am; therefore, determine the type of product that I might be interested in. Thus, leading to targeted advertising. I think there is an opportunity here for Congress as well as the tech sector to think about ways to look at how to correct bias. Are there secondary datasets that they can use to sort of ensure, as Mr. Luckie has said, that this algorithm is not going to generate an unintended consequence? Are there cases where they want demographic data and want permission from consumers to collect that, to ensure that the algorithm will not be biased? I think as we go forward those conversations will need to be had. I think it is also important for customers and consumers to have a feedback loop. Ms. Kelly. Let me just get my last question in---- Dr. Turner Lee. Yes. Ms. Kelly [continuing]. Which you can answer and anybody else. What is the role of the Federal Government? Do Federal laws and guidelines need to be updated to reflect changes in advertising technologies? So, I just wanted to quickly get that. Dr. Turner Lee. Yes, and I will be quick. I have said it, and I will keep saying it. I think we need to revisit those nondiscrimination laws and, where they are applicable, apply them to digital space, and maybe not do it in a way that is punitive, but just extend those protections to consumers. Everything that you have heard around algorithm bias is mitigated through existing guardrails, but I also think that it is important that we have self-regulatory measures where the tech sector sits down with civil society, it has been mentioned, to think through these cases. There are going to be use cases where data will be weaponized against communities of color, against women, against people with disabilities, and we need to find ways to stop that. Ms. Kelly. Thank you. Ms. Schakowsky. Now I recognize Congressman Gianforte for 5 minutes. Mr. Gianforte. Thank you, Madam Chair. And thank you for the panel today, for your testimony. This is a very important topic. In our technology business, we found that internship programs and coding classes were effective ways to train and recruit good hires. And I would love to hear from Dr. Ferrini- Mundy. With that in mind, can you please discuss some of the steps your university is taking to engage with the local business community to match up the educational pursuits with the needs in the marketplace? Dr. Ferrini-Mundy. Thank you for the question. We, of course, are situated in Maine, in a rural State like yours. So, we are very, very eager to be certain that we are serving the economic development of the State of Maine with very well-prepared students who will engage that business and industry. And there are a few key fields where this is especially opportune for us in Maine, in the forest resources industry and agriculture fields, in the marine science areas, for a few. Those fields are all becoming increasingly technological. So, they are not high-tech in the sense that we are discussing here exactly, but they really do depend upon people who will bring the kind of knowledge that we are discussing. So, we are very eager. We have a number of important internship opportunities. We have an incubator that allows students to work together with companies that are looking to expand and new companies coming into the State. We try to make those real-world problems that these companies are facing a part of the education of our students. Mr. Gianforte. OK. Thank you. We have been talking a lot today about the shortage of workforce. One of the things we have experimented with--not experiment, we have done it in Montana--was we, particularly in the computer science curriculum, we have introduced a bachelor of arts in computer science versus a bachelor of science---- Dr. Ferrini-Mundy. Right. Mr. Gianforte [continuing]. To attract more people. Mr. Guthrie was asking about how do we broaden the net to attract more people into these STEM programs at a college level. Could you just talk to that a little bit, about other things we might try? Dr. Ferrini-Mundy. Sure. And I should say that, prior to coming to the University of Maine, I worked at the National Science Foundation for a number of years, and was a part of a variety of conversations there. In part, the NSF has identified something called The Future of Work at the Human-Technology Frontier, as one of its initiatives that is described on their website. And that is a program that is calling for research that will help us to better understand these technological changes that our society is addressing and how we can really better understand what it takes to prepare people to work in these spaces. So, within computer science, for example, there is a national conversation about what you described, creating the BA in computer science, but also the notion of computer science plus some other field as a kind of major. So, computer science plus biomedical engineering, computer science plus sociology. The idea that we want to advertise to students, the computer sciences is meant to help us solve a very wide range of problems, not only problems that are specifically in some vision of technology that may be an old-fashioned one. Just one quick last point that I wanted to make relative to the discussion of the algorithms and the algorithmic bias. I do think that a piece of addressing this should be sitting within universities as well, so that researchers are working within machine learning, within AI, to understand and help to shape these algorithms in ways that are consistent with the kind of diversity we are talking about. Mr. Gianforte. Well, I certainly agree with you. It is a truism that computers are here to stay, and I can't imagine any degree that wouldn't benefit---- Dr. Ferrini-Mundy. Right. Mr. Gianforte [continuing]. Without some minor in computer science---- Dr. Ferrini-Mundy. Right. Mr. Gianforte [continuing]. Making a better candidate for any job in the marketplace. Dr. Ferrini-Mundy. Absolutely. Mr. Gianforte. Yes. Ms. Oliverio, I understand your organization works with veterans. We have the second-highest per capita number of veterans of any State in the country in Montana. Some people might have a difficult time understanding how a ranch hand or a combat veteran might end up in the tech industry, but I believe the key to a good employee is always work ethic and selflessness. I am interested in having you just summarize for me briefly the work that you do to help veterans make that transition into the tech industry. Ms. Oliverio. For us at Military Talent Partners, we believe that everything begins with mentorship, understanding the goals and the purpose and helping veterans and spouses really find their ``why'' and understand what they want to accomplish in their career. It may have absolutely nothing to do with their job in their service, but by aligning their goals and their purpose, they become empowered to find a meaningful career that is meant for them. Mr. Gianforte. OK. Great. And just, Ms. Houghton, if I could, I understand some companies exclude people with disabilities. One of the reasons they give is the cost. Can you just comment briefly on how can companies accommodate costs associated with hiring people with disabilities and what impact that really has? Ms. Houghton. So, I think that that is a myth, and that what we have found and what we have shown with the Accenture data is that it is quite the contrary. For companies that are committed to disability inclusion, they actually four times greater total shareholder returns. Mr. Gianforte. Thank you for your enthusiasm. With that, I yield back. Ms. Schakowsky. Mr. O'Halleran, I recognize you for 5 minutes. Mr. O'Halleran. Thank you, Madam Chair. And, Panel, I have really appreciated this discussion today. Mr. Luckie had identified within his written testimony at least that there were 22 percent of rural residents that did not have the high-speed broadband ability to be able to even get prepared for the industry. Whether it is gender bias or racial bias or disability bias or geographic bias, and it really bothers a lot of us on all those fronts, along with our veterans, the bottom line is that it is still human input into this process and somewhere along the line these organizations need to identify that peer review, and input from the community is critically important to be able to get some fairness into the process. I would rather see that done through the companies themselves, and I think that that is the direction we need to go. I am going to kind of go to the geographical issue right now. Dr. Ferrini-Mundy, you have mentioned a ton of programs at the university, but it gets down to how do we get students from those geographic areas. My district in Arizona has 12 Native American tribes that make up 24 percent of the district's population. Twenty-some percent of the district population is Hispanic. A tremendous amount of poverty. Arizona is a big State. Sixty percent of the land mass of that State is in my district. And so, the problem becomes that, when it gets down to just the sheer technology needs, and now that we are going from where we are at now to 5G, I think personally that that is going to put rural America and those areas back further, even though they will get better than what they have, they will lose ground from where people are going to be in urban areas. What is your experience in getting students from those areas, first, into the university, keeping them in the university, and getting them into these programs? Also, we have, obviously, many of those families that the first time they hit the university is the first person from that family ever to get there. Dr. Ferrini-Mundy. Right, right. So, I can speak to that in a couple of ways. In the University of Maine, we are actually a part of a university system with several regional campuses across the State, many of them located in extremely rural areas. In fact, one of them, the University of Maine at Machias, is a regional campus of the University of Maine. And so, we are very dependent upon making certain that those campuses, and particularly Machias, are responding to the challenges and issues of that particular geography. It is a coastal area. It is a very economically disadvantaged area. So, we are seeing that campus really thrive as a center for the community, as a way to address issues that are of interest there, to try to build a workforce that can thrive in such areas as healthcare and community services. So, some of this is about customizing what the institutions offer to the regions that are there. The broadband issue is a serious challenge in Maine. And so, we face that in a variety of ways, working together with the State, with the legislature, to see what kind of progress we can make on that front. Because online opportunities will continue to abound, and we want to be certain that those are accessible to all students. Mr. O'Halleran. How much of a disadvantage is this to those residents and their children from those areas versus urban areas? Dr. Ferrini-Mundy. It certainly is a major challenge. That is why at this point we want to be certain that our regional campuses are providing very good opportunities and services, ranging from programs for first-generation college-going students to other kinds of support. But it is a serious problem for us in our State. Mr. O'Halleran. And just a comment towards the end here on our veterans. First of all, it is Women's Month and we need to make sure that--we can't just take 50-some percent of our population and not them have an active high-level role in our society and leaders of our society. But our veterans, again, when you take a look at the training that our service personnel go through, that is a key indicator of leadership in the future, the ability. All they need is that little bit extra to be able to adapt. It is not changing; it is adapting to a new role. We all go through that in life, and we just need to make sure our veterans have that opportunity. So, thank you for what you are doing. I yield. Ms. Schakowsky. Mr. Carter, you are recognized for 5 minutes. Mr. Carter. Thank you, Madam Chairman. And thank all of you for being here. We appreciate this. This is certainly a very important subject. I want to start with you, Ms. Ferrini-Mundy. Is that right? Dr. Ferrini-Mundy. Yes. Mr. Carter. I'm sorry. Thank you. Obviously, you have got an extensive background in developing policies that would help young people enter into the job market and into new opportunities, particularly as it relates to STEM and particularly as it relates to getting a number of minorities involved, a number of those who don't have the opportunities perhaps that some others do. Would you agree that there is a major problem in the number of minority communities that have access to these STEM fields and tech-related jobs? Dr. Ferrini-Mundy. Certainly if we look at the numbers of minorities well represented in the STEM fields, there is a problem. Mr. Carter. Right. Dr. Ferrini-Mundy. I mean, those numbers should be tracking at least with representation across the society. Mr. Carter. So, let's talk about that for just a second. When you say ``STEM,'' I think we all think of just STEM and more engineering. Dr. Ferrini-Mundy. Right. Mr. Carter. But there is more to it than that. Dr. Ferrini-Mundy. Certainly, certainly. Mr. Carter. In the State of Georgia, we have been very successful in building up our film industry through tax credits and different incentives, and through the work of the Georgia State legislature, and particularly the economic development committee in the House and our chairman. Chairman Ron Stephens has done an outstanding job. It has resulted in a lot of opportunities for these type of jobs. It has created a number of jobs. One of the things that we are very proud of is the Georgia Film Academy. That was established through the State, and it runs through 12 different institutions, through the university system and the technical college system. They offer degrees and certificates, and that is the type of thing. Can you think of any other examples like this where it necessarily might not be engineering jobs per se, which is I think what we think of when talk about STEM? But it is opportunities, nonetheless, and good opportunities and good- paying jobs. Dr. Ferrini-Mundy. I would mention a few areas at least that are relevant for us along these lines in Maine and that do require some combination of background in science, technology, engineering, and mathematics. I will keep putting in plugs for mathematics and statistics as central features for these areas. But industries such as the forestry industry in the State of Maine or aquaculture, a growing industry in our State, these are industries that do depend on technological solutions to some of the challenges that they face. Agriculture, more generally, is, of course, also facing opportunities with new technologies. So, people with a range of backgrounds really can come together to solve the sorts of problems that these industries face in States that are particularly suited to particular industries. Mr. Carter. Great. And thank you for mentioning timber because Georgia is the No. 1 forestry State in the Nation. That is something that is very important to us as well. Real quickly, I am going to switch over to you, Ms. Oliverio. I'm sorry. I hope that is OK. But I am very blessed; in the First Congressional District of Georgia we have four military installations. We have a number of veterans. We are home to over 75,000 veterans. Our quality of life and all the things that we enjoy, our environment, lead to a lot of people retiring in our area, particularly after we have four military installations, a lot of them just stay there and retire. The hiring of veterans, this is something that is very important. A lot of companies in the First District have really capitalized on this and found the veterans to be excellent employees. What are some of the challenges that we face, that veterans have faced, to being hired? Ms. Oliverio. Specifically, to the major Atlanta area, there are a number, a massive number of veterans that want to relocate to that area and to work, and have struggled to find meaningful job opportunities to get in front of the interviews. So, while there is a huge footprint for veterans in that area in the beautiful State of Georgia, we can do better on the messaging on what is available for veterans and how we can better acclimate them into the businesses in that State, and then, therefore, retain them and mitigate turnover issues across the landscape of the corporate build in Georgia, and in specifically Atlanta. But if we can make it more well known as to what is available and welcome them in, we would have a much higher success rate. Mr. Carter. Is there anything we can do from a State perspective or from a government perspective as far as certificates, as far as our technical schools are concerned? Ms. Oliverio. Absolutely. I think encouraging mentorships or fellowships or apprenticeships of technical schools is an excellent idea. It is making it known. We create a lot of great programs and ideas, but the word doesn't get out, and it becomes too difficult for people to understand how they can apply to it or how they qualify or how their business will fit into that model. So, by making it more accessible, they can take more advantage. Mr. Carter. Great. Well, thank you all very much. This is extremely important. And thank you, Madam Chair, and I yield back. Ms. Schakowsky. Thank you. And now, I am calling on Congresswoman Blunt Rochester for 5 minutes. Ms. Blunt Rochester. Thank you, Madam Chairwoman, and thank you for having this very important panel. Diversifying and including folks in the tech industry is one that I think all of us on the committee can agree with. I really wanted to use this time to talk about the fact that, whether it is diversifying the workforce or whether it is the products that are being created, if you don't have representation, we don't get a good product. We have already heard the stories of facial recognition that might not recognize darker-skinned people or voice-enabled devices that have a hard time with accents. And at its worst, we have seen algorithmic bias result in criminal risk assessment software predicting that black offenders were almost twice as likely to reoffend than white offenders, even though their history showed otherwise. So, it is really vitally important to have this panel. I am grateful to the diversity even of the panel that is represented here and for the work that you do. My first question is for Dr. Turner Lee. In your written testimony, you mention COMPAS, the AI software used across the country by judges to determine how likely a convicted criminal is to commit more crimes. Can you speak more about the real- world consequences of the bias found in the program? Dr. Turner Lee. Thank you, Congresswoman. Yes. I think as a researcher--and I have a degree in sociology--I think that is a really important use case that deserves some more discussion. What we are talking about in the COMPAS algorithm is the ability of judges to rely upon automated decisionmaking to assess whether or not people should have longer sentences, be released on bail, et cetera. But the challenge with that comes--and I think we have heard it a few times--in the training data. We all know in the criminal justice system that African Americans are more likely to be arrested. They, therefore, are more likely to be populated within this training set. As a result of that, when it comes to sentencing, they are more likely to be penalized or assessed larger penalties because of their overrepresentation there. That is problematic because what that is suggesting is that we are not taking technology and assessing against settled research, settled stereotypes, or information that we know are assumptions about stereotypes, which brings in, I think, what was discussed around implicit and conscious bias, and it also brings in structural race and the discrimination. Unless we fix those, we, then, develop products that will continue to pick up on those errors. Ms. Blunt Rochester. Right. Dr. Turner Lee. And so, again, Congressman, you picked up on something that is quite problematic because there are irreparable consequences to being incarcerated longer that we cannot solve. Ms. Blunt Rochester. Thank you. And it leads me to Mr. Lopez. I know you were the longest- serving general counsel at the U.S. EEOC. What role should the EEOC play in all of this? And do you have the tools? Mr. Lopez. Yes. No, I think the EEOC has been very active in terms of collecting the data on the absence of diversity in the high-tech field. It has also held, I believe, three commission meetings which focused on developing information on the use of big datasets as employment screens, which is, I think, one of the more controversial and one of the most important areas in terms of how predictive analytics are now being used as a way to recruit and to select applicants. And I think what happens is that many of the companies involved in these efforts market them as a way to actually eliminate bias, and that is a possibility. I mean, really big data, data analytics, if used correctly, is not necessarily a bad thing. I don't think any of us are coming here as Neo- Luddites against the technology. But I think the EEOC has been very active in terms of talking about the problem of bad data in, bad data out, some of the inherent biases in these tools in terms of addressing discrimination. Ms. Blunt Rochester. Thank you. And, Ms. Houghton, I want to briefly say that I was pleased to see the work that you are doing, and I also wanted to reinforce what you said about the statistics for people with disabilities. When you intersect that with being a woman or being a person of color, it is even double. In the last 20 seconds, could you talk about some of the myths that people, particularly employers, have for hiring individuals with disabilities? Ms. Houghton. I think that that is a great question, Congresswoman. And there are a lot of myths. There are myths that the talent doesn't exist. There are myths that, if I hire these people, that they are going to cost me money, that they are going to sue me, and on and on and on. Ms. Blunt Rochester. Great. And I yield the balance of my time. Ms. Schakowsky. Mr. Bucshon, you are recognized for 5 minutes. Mr. Bucshon. Thank you very much. My daughter is currently majoring in computer science at Cornell University. She is a junior, and she is primarily in STEM because, when I first came to Congress, I heard all these stories about how we needed more diversity in STEM, which is very true, and we all agree on that, I think. I told her, look, if you want a job, you should probably look at computer science, and she did. So, it is the only kid that has listened to me so far. The rest of them, you know, they won't listen. [Laughter.] And she will be interning in a large bank in New York City this summer. And so, in this discussion I think--and I apologize, I had another hearing--but we a lot of times focus on edge providers like Facebook, Google, Twitter, and other tech companies like that. But, really, these opportunities are growing across all sectors of our economy. And so, Dr. Ferrini-Mundy, can you explain how universities are working to promote these types of tech opportunities to their students, including, for example, the big banks which are hiring more tech people than they are bankers? Dr. Ferrini-Mundy. So, again, I think there are many things universities are doing and can do. One is to make sure that the education that we provide our students both deep and broad enough to enable them to seek those kinds of opportunities. The second is to continue to work to have internships, apprenticeships, opportunities for students along the way as undergraduates to get access to chances to work in some of these different sectors, and to both show what they bring because of their diversity and their knowledge, and also get a chance to imagine whether they could work there. Those frequently turn into offers at those kinds of companies. Mr. Bucshon. Based on American Community Survey-PUMS data, 1.1 percent of people from Indiana have a degree in computer science. Again to you, diversity is important in all industries, including diversity of ideas and background. Can you provide examples of what efforts universities might be taking to promote STEM degrees like computer science in rural communities? And honestly, I think this may be as applicable to urban settings also, but particularly rural communities that may not have the knowledge of these job opportunities offered in the tech sphere. Dr. Ferrini-Mundy. A lot of the examples that I am aware of have to do with working at the K-12 level, to actually go into schools, to create after-school learning opportunities, coding kinds of programs. So that students can get a picture of what it might be like, rural students particularly who may not have a good opportunity to see these options, and then, imagine what they could be in those careers. So that they can follow a pathway that will take them toward an undergraduate major. Mr. Bucshon. Dr. Turner Lee, why do you think we are having a discussion about the lack of diversity in tech? I mean, what is your fundamental view about--I mean, why? I think we have answered some of that today and we have given some opinions today. But that is fundamentally why we are here, right? We all believe that diversity is very important, diversity based on like my daughter; she is a woman. And then, what is your view? I mean, fundamentally, when you get really cut to the chase? And what can we do about it? Dr. Turner Lee. Yes, I think, fundamentally, we are having this conversation because we are in this fourth Industrial Revolution, right? And this Industrial Revolution has people tethered to technology, and we are seeing the consumption trends sort of amplify itself to the extent to which people who are now consuming these products are not part of the decisionmaking of these products. And just like we saw in other industries, let's just go back, we saw in the '60s and '70s the same type of diversity programs initiated to bring in more representation, and they made companies better. And so, I think, going forward, we are-- -- Mr. Bucshon. Yes, for medicine, for example. Dr. Turner Lee. Yes, medicine. Mr. Bucshon. I am a doctor. My wife is a doctor. Dr. Turner Lee. That is right. Automobiles---- Mr. Bucshon. Right. If you look back at the '60s, the makeup of people who are physicians was not very diverse, right? Dr. Turner Lee. Yes. I just think we are at a turning point, Congressman. I think it is one of those conversations that for years people have been discussing, but, as technology becomes more entrenched and the boundaries between the physical space and the digital space become much more connected, that we are just trying to figure out as a country, for us to be internationally competitive, how do we bring in more diverse perspectives and minds. That is why the diversity of this panel reflects this conversation. By the same token, we don't want these products that people are depending upon--we are no longer an in-line economy; we are in an online economy--to hurt them or harm them. And that is why we are having a discussion around consumer protection. So, I think that is at the crux of why we are here today. Mr. Bucshon. OK. Thank you. I yield back. Ms. Schakowsky. Beautiful. I am calling on Mr. McNerney right now for 5 minutes. And I just want to say, you are seeing that people are coming back. They really feel that this panel and this discussion is very important. Mr. McNerney? Mr. McNerney. Well, thank you. I thank the Chair. And I thank the witnesses. I apologize for missing most of your testimony. But, first of all, I want to say I am a mathematician by education. I cochair the Congressional AI Caucus. And I represent a district that is very, very diverse. So, while I am excited about many of the technology innovations that we are witnessing today, I am also concerned that many people will be left behind, and what that means for my district and my constituents. Specifically, I am worried about how automated decisionmaking can impact my constituents' ability to obtain a loan, to receive social services, to see housing ads, to be promoted in jobs, or even to get consideration for a job in the first place. So, it is clearly critical that the teams designing these products are representative of who the products are going to serve. Now there is a real need for greater transparency in how these algorithms are produced and the decisionmaking is made. Mr. Lopez, in your written testimony you noted that algorithms are often predicated on data that amplifies, rather than reduces, the already present biases in society, racial, ethnic, and socioeconomic. Can you elaborate on that a little bit, please, and maybe provide an example? Mr. Lopez. I mean, I think a good example would be using social media to advertise for housing opportunities and to limit the advertisement to certain zip codes. Zip codes have traditionally, due to the history of housing segregation in this country, have often been used as a proxy for minority---- Mr. McNerney. So, is that being done by algorithms or by human beings, those decisions, or both? Mr. Lopez. Everything that we are talking about is being done by human beings at the end of the day. I think what happens is that the algorithms--and let me use a different example from the employment context. Let's say there is a correlation between having an interest in manga comic books and being somebody who might be a good computer engineer. This is a real example. But, again, the individuals--it is correlation, not causation--and the individuals that tend to have an interest in manga comic books tend to be men, right? And so, if you start to use that algorithm to select computer engineers, what you do is you sort of reinforce and you replicate, I think, the general systemic exclusion of women. Mr. McNerney. So, that is an example of why machine learning makes bias more difficult, right? Mr. Lopez. Yes, yes. Mr. McNerney. Thank you. Ms. Turner Lee, in your testimony you mentioned that you will be issuing a paper next month addressing a course of developing quality instruments and questions to measure algorithmic bias. Do some of these instruments already exist? Dr. Turner Lee. There are some instruments, and I think it has been mentioned that we have seen companies more on the tail end that do audits. The paper that we will be releasing at Brookings is more of a forward-thinking paper, to your point, and it is coming up with questions. Do we need an algorithm for this actual use case? If we do, is there diversity in design? A lot of the questions that you are asking. Are the right people at the table? Is civil society part of the conversation and debate? Is there a feedback loop? I think that is part of the pathway toward more responsible governance over the way that we are constructing algorithms, evaluating, et cetera. But I do want to suggest to you, Congressman, something in your previous question. I don't want us to put in a bucket that all technologies are acting discriminatory or racist. Mr. McNerney. Right. Dr. Turner Lee. I think the way the technology has been structured and how opaque the internet has become, these algorithms are adapting to our human behavior. So, there are cases where a developer, a technologist, may not say, ``I want to flag women for this particular job.'' But how the internet has actually become layered, it will see the name Mary and assume that Mary is a woman, and populate itself and operate and function pretty much adapting to the historical biases that we have as a society. Mr. McNerney. Again, that is machine learning, a machine- learning tool. Dr. Turner Lee. That is machine learning, yes. And I think the paper that we are going to be putting out is really trying to help companies have a more proactive stance to actually how you look at these conditions and how you look at these teams, and how you look at these outputs, and come up with solutions to do something about it. Mr. McNerney. Well, you also mention that businesses have taken action to correct bias, such as Facebook and Google, but that was only after a lot of public pressure. Are there examples of companies that have proactively acted or do you think that is a trend that we could expect to see without significant motivation from Congress? Dr. Turner Lee. I actually want to say that there are companies that are taking advantage of this. Microsoft is another case where they are actually very proactive in how they design algorithms. They had a case where they put out a voice- activated computation or application that was picking up--it was taking in people's voices and, basically, putting out very antisemitic and racist stuff. They took it off market, right, because the developer did not anticipate those consequences. So, I think we are seeing a blend, which is why I said previously and put it in my testimony, self-regulation is not a bad idea. It is just a combination of how do we combine that framework with what we already have existing in our resource kit. Mr. McNerney. Thank you. I yield back. Ms. Schakowsky. Mr. Soto, for 5 minutes. Welcome back. Mr. Soto. Thank you, Madam Chair. It was briefly mentioned before, a report came out from the ACLU about new facial recognition technology where they downloaded 25,000 arrest records, used them against pictures of every current Member of Congress in the last term. There are 28 false matches. People of color made up 20 percent of Congress at that time, more now, by the way. And 40 percent of the false matches were people of color, including legendary civil rights hero John Lewis. Obviously, the software as it stood there would disproportionately target minorities. This is a technology that is being used in my hometown of Orlando, only voluntarily, to track officers to test the technology, but certainly it is something that is concerning for us. Recently, I got to join Representatives Brenda Lawrence, Ro Khanna, and others, on ethical guidelines for AI development, including transparency and process, empowering women and underrepresented populations, and accountability. So, it really brings up sort of a broader topic of what Congress' role is in being able to address these things. I will start with Dr. Turner. Some of the algorithmic bias mentioned today is the result of bias datasets, are there practices and procedures that can be implemented to reduce the bias in training data that could be helpful? Dr. Turner Lee. Yes, I would say start with overrepresentation, right, of what those faces look like. So, an MIT researcher--and I don't want to butcher her last name, maybe Mr. Luckie can help me--Joy Buolamwini has actually done a lot of studies where she has actually said a lot of facial recognition technologies do not work because they are underrepresented in terms of African Americans or darker skin hues. So, we have to actually populate the training data where it reflects the entire population. I think part of the problem we have, Congressman, why people don't do that, as a researcher, I am subjected to guidelines when it comes to IRB requirements and human conditions. We are rushed to market when we come to proprietary algorithms, right? It is who gets there first. And so, I think having some attention to overrepresentation is really important. Also, feedback loops also help with the training data. The paper that we will be releasing will talk a little bit about the technical side. Again, I'm not a computer scientist. I am interested in the civil rights outputs that actually come as a result of that and what legislators should actually be looking for. But I think combining those conversations and having multi-stakeholder conversations is also helpful because the left hand often doesn't know what the right hand is doing. Mr. Soto. Thank you. And for Ms. Kim, what is the role for increased transparency and explainability in reducing algorithmic bias? Ms. Kim. Thank you, Congressman. In looking at the decisionmakers within the tech sector, the employees, the professionals, the technicians, as well as the executives that make much of the decisions, we need to have more diversity. And the transparency that we need is more data. We applaud the efforts of companies that voluntarily release diversity data for recruitment, retention, and attrition data. But, for the AAPI community, that data is incomplete because it is not disaggregated. Our communities are so diverse, and we have to look beyond the glare of the model minority and say, what is actually going on behind the aggregated number? Yes, there are more Asian tech workers than other minority groups, but, in fact, if you look deeper--and we don't have these numbers, but based on other industries and other Census data and other information, we know that there are communities that have incredibly high poverty rates, low educational attainment levels, and high limited English proficiency levels that don't get represented at educational levels in schools, in other sectors. And so, we need more data and more information and transparency from tech companies, so that we can see what is actually going on underneath that model minority myth of Asian-Americans are doing OK. Mr. Soto. Thank you, Ms. Kim. And I just want to end with sort of a personalized story from my own family. My little cousins, I can't get them off of video games. They are of Puerto Rican descent, like I am. And it turned out that when one of my cousins went into the Air Force, that skill set ended up serving him well to be one of only two people out of two dozen to actually pass the drone aviation course. And it occurs to me how critical having access to technology at an early age is. When you look at Bill Gates or Bill Joy, or others, the stories go about how they had access to computers early on, and that proved critical in them getting their 10,000 hours ahead of everybody and being able to really change the world. So, those are things that we also will be taking very seriously to get into, access at an early age for young people of all communities. Thank you, and I yield back. Ms. Schakowsky. So, I want to thank our panel, but I also want to see--I spoke to our ranking member. I would like 5 more minutes, and I would welcome her taking 5 more minutes. Mr. Luckie, we have talked a lot about the pipeline and making sure of the educational system and educational opportunities. But you said something that really struck me. It is that, even where women and minorities, and others that aren't represented in the tech industry, are available, that they still are not hired for those jobs and advancing in the positions. We also see women leaving those jobs earlier than men. So, rather than blame the victims, you know, you have got to go to school and you have got to take these courses, what are we going to do about--there is obviously some discrimination. I really want to stop here, but ask the rest of the panel, what should we be doing? You have mentioned, Dr. Turner Lee, that there are discrimination laws that are in place. You have talked about the EEOC, Dean Lopez. I mean, what should we be doing, both about the algorithmic bias that is there, but also just about hiring people, investing in people right now? Mr. Luckie. Well, it is like you said, a lot of the onus has been put on the people who are graduating from STEM or who have not heard about STEM to break into these tech companies. But what we are finding is that the stopgap is actually happening within the interview process, where employees are not seeing people who look like them. And so, they are choosing the people who look like them, come from the same backgrounds, come from the same schools, which puts others at a disadvantage. We are seeing it in the discrimination that people face once they are inside of the companies. What is happening is that people are graduating from STEM, but end up choosing other industries because they see the discrimination. I think the most important point that we should take away from this is that the people who come from diverse backgrounds are the representatives of these companies who go back out into the communities and say, yes, you should be STEM, be in the STEM areas. And so, we have to call on those people to say, hey, we need you to go back to the communities; the people who are coming from Maine and from Texas, and from these locations, to go back and say, you can do this because I am inside of the company now and I am making it work. Ms. Schakowsky. I wonder, Dr. Turner Lee, how can we use the current laws to help here? Dr. Turner Lee. Yes, I think we have some tools in our toolkit. A lot of the stories that we have heard today, it is not just the one-time action where it happen, but it is this latter consequence. The former Obama administration was real key about putting out reports on algorithmic bias. It is not a new topic to us. The difference is, as we go over time and we let this evolve, that we are going to find people being denied loans. And I don't think we want to see structural discrimination actually find its place into the internet, which has always been the low-hanging fruit for opportunity. So, I think what Congress does have at its disposal are tools to have a conversation around nondiscrimination and to see which ways do we extend that. And companies, I think, in the tech sector are starting to realize that. Facebook, in particular, last year pulled down a process that was discriminating in the case of housing, where advertisers were able to click off who they did not want to serve, and they immediately stopped that process. But they didn't realize that the Fair Housing Act was one of the levers for why you shouldn't do that in the first place. So, I think Congress has an opportunity to have conversations about that, as we have had in the past, as we see these transitions happen. And I also think it is important, just to complement your previous question, Congresswoman, STEM has to be a national priority. This is no longer a conversation of investing in programs. We have to invest in people. And if we invest in people, we will have an international competitive edge on any of these careers, particularly as digital access becomes much more permeated and much more embedded in our society. Ms. Schakowsky. There is almost a minute left. Anybody else want to comment on that? Dr. Lopez? Go ahead. Mr. Lopez. Congresswoman, as part of Congress' oversight responsibilities over the agencies, I think there is a real opportunity to ask the relevant agencies, EEOC, HUD, Department of Justice, FTC, anybody operating in this space, what they are doing in this area. And it doesn't, obviously, have to be hostile. I came out of the EEOC. I happen to think that they have been very forward-thinking in this area. They continue to be forward-thinking. But I think that the oversight responsibility and the power of the purse is very important. Ms. Schakowsky. Go ahead. And did you say something? Anybody else? OK. Mr. Luckie. I would just like to say, in terms of the larger conversation, that we have to stop treating AI and algorithms as omnipotent, as if they know everything. They are still being programmed by humans, and we need to recognize that in order to curb this bias. Ms. Schakowsky. Thank you very much. Mrs. Rodgers. OK. Madam Chair, I wanted just to follow up on your line of questioning right there. As I said in my opening statement and opening remarks, part of the challenge here is that it does demand change. It is going to demand change in our culture. A lot of the focus today was on recruitment, and I think that is an important piece of this. We have to focus on the pipeline. That is very foundational. Education is important, skills training, access, and exposure to what is possible. Helping people imagine is important. We have talked a lot about the value of teams and having teams--Mr. Lopez, in your testimony it says, you highlighted that teams that are made up of individuals of diverse backgrounds are more innovative and generally make more error- free decisions. I, too, just wanted to take this at the close here to kind of those next steps. So, even after we get some of these individuals hired, what is working as far as the retention and the promotion? Because there are these next steps. We have to do better at retaining these employees, and then, promoting them to leadership positions. I know I have seen some of the work on women where women respond to the coaching and to giving that feedback. But, often, when they are given a chance for leadership, they shine. So, I, too, just wanted to open this up at the end and ask, what do you see working when it comes to the retention, and then, the promotion to leadership? And how do we better invest in these people after they are hired? Do you want to start it, Mr. Luckie? Mr. Luckie. Sure. So, I will say the best thing that is driving retention is the employee resource groups that are in major tech companies and businesses everywhere, where women are helping women, black people are helping black people. Part of the reason why I stayed at Facebook as long as I did is because of the black ad group, which was the employees that were coming together, mentoring each other, exploring other opportunities inside the company. The more you empower those employee resource groups, give them the budget, give them the space in which to operate, the more you are going to see that retention increase within these companies. Mrs. Rodgers. Thank you. OK. And so, do you want to---- Dr. Ferrini-Mundy. Yes, I will just add to that. What Mr. Luckie is explaining has to do with cultures within these environments, and those cultures get built by the people who are there. And so, I would still make my argument that education and the pathways that get provided through education, which include early opportunities to work together in groups, to learn how to be resourceful within particular subgroups, all of that is critical to building that kind of culture. Mrs. Rodgers. All right. Thank you. Ms. Kim or whoever is next? Yes? Dr. Turner Lee. If I can, in a previous life I worked with the cable industry doing some work. What I found there is that, internally, companies have to have metrics, they have to have accountability, and they also have to have some type of executive support. Without it, it doesn't work. Where we know diversity and inclusion works is when there are bonuses tied to leadership. We know that it works when the executive claims that this is the only way we are going to actually conduct business, training, et cetera. And those invested resources happen at an internal level. Obviously, we can't manage what companies do inside, but what we can do--and this is something I think in terms of what is next--is we can place a level of stewardship and responsibility on companies through their reporting, whether it is mandatory or voluntary. There are tons of regulated industries that provide voluntary information and scorecards on how they are doing with diversity. And that is something I think is going to be an acceptable practice going forward, because we, as consumers, want to know that people are doing the right thing. And so, I think, going forward, it is going to take a mix of the internal pressure and the external pressure to actually move to a place of, I want to say peace, where everybody can just get along, but where we actually indulge ourselves in diversity in ways that make us more profitability and much more successful as a country. Mrs. Rodgers. Thank you. Do you have any thoughts? Ms. Oliverio. Yes, thank you. I spent more than 10 years in recruiting all across technical fields, across the U.S. And in any company, in any demographic, in any background, the key to retention is making employees feel like they belong. Much like Mr. Luckie was stating, employee resource groups are amazing, but the core value there is making employees feel valued, making them feel like they belong. That is where veterans and military spouses struggle most to feel like they are a part of a unit again, to feel like they are welcomed in their work environment, that they have reasonable accommodation, and that they belong there. There, they are more likely to stay and to be happy and to do well. Mrs. Rodgers. Thank you. Ms. Houghton. In addition to seeing yourself and feeling a sense of belonging, I think one key thing that I want to make sure gets on the record is that, internally, within these companies we need accessible technology. If people are going to grow, technology accessibility has to be built in from the beginning, not as an afterthought. Mrs. Rodgers. OK. Thank you, everyone. Ms. Schakowsky. Thank you so much. I think this has been a terrific panel and something that we, as a Congress, need to follow up on now. Thank you. I request unanimous consent to enter the following documents into the record. Where are they? OK. The National Urban League letter on diversity in tech; a letter from the Electronic Privacy Information Center; the Internet Association's letter for the record; testimony from Jennifer Huddleston, research fellow at the Mercatus Center at George Mason University; a letter from Maxine Williams, Facebook's chief diversity officer; statement from Representative Maxine Waters; a blog post and case study from Charter Communications. Is that it? Without objection, so ordered. [The information appears at the conclusion of the hearing.] Ms. Schakowsky. Oh, I have one more page? This? OK. I would like to thank the witnesses for their participation in today's hearing. I remind Members that, pursuant to committee rules, they have 10 business days to submit additional questions for the record, to be answered by the witnesses who have appeared. I ask each witness to respond promptly to any such question that you may receive, and you may receive them because a number of people were traveling from different hearings. So, at this time, the subcommittee is adjourned. [Whereupon, at 12:46 p.m., the subcommittee was adjourned.] [Material submitted for inclusion in the record follows:] [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT] [all]