Discrimination Should Not Be Automated: Tell Congress to Pass the Algorithmic Accountability Act

From Color of Change (https://colorofchange.org/):

Companies are giving more life-changing decisions over to algorithms—computer systems that can solve problems or complete tasks—reinforcing the discrimination that Black communities fought hard to outlaw in offline spaces.¹ Humans introduce racial biases to algorithms through various points; an example of this is training algorithms using datasets that reflect the prejudices and inequalities of our society.² That’s why algorithms used by car insurance companies may determine higher insurance rates for people living in predominantly Black neighborhoods.³ Algorithms can also exclude women from seeing certain employment ads, reject qualified Black students for school admissions, or deprioritize Black patients for care.^4-6 Algorithmic accountability is the only proven method to prevent algorithmic bias, and that requires evaluating algorithms, deleting discriminatory datasets, and suspending the use of discriminatory algorithms.⁷ We need Congress to pass the Algorithmic Accountability Act to protect us from algorithmic discrimination in important life decisions like securing housing, education, employment, or obtaining credit.⁸

Tell Congress to Pass the Algorithmic Accountability Act at https://act.colorofchange.org/sign/algorithmic_accountability.

Crystal Marie and Eskias McDanielses’ difficulty in obtaining a housing loan is a real-life example of how algorithmic bias harms Black people. The McDanielses were ready to buy a house in Charlotte, North Carolina, and were more than qualified to receive a housing loan due to their excellent credit scores and high household income. But after shelling out $6,000 in fees and deposits, the lender denied this Black couple a home loan.⁹ The couple wasted thousands of dollars because the lender used a racist algorithm to make this unacceptable decision. Fortunately for the McDanielses, they were eventually able to get the house after finding out an algorithm was the source of their initial denial and challenging the denial. But not everyone has the time or resources to challenge these harmful outcomes or even know they’re happening. It’s illegal for lenders and other decision makers to discriminate based on race, therefore the use of discriminatory algorithms by these folks must also be outlawed.

While the McDanielses had to fight an unjust algorithm to get a housing loan for their house, access to such resources should not be a fight at all. The Algorithmic Accountability Act will address algorithmic biases by:

Requiring assessments that will help protect Black people from automated discrimination

Targeting decisions made in the context of housing, healthcare, education, employment, credit, and insurance

Giving the Federal Trade Commission (FTC) the authority and staff necessary to require these assessments

A better future is possible—a future that protects Black people from algorithmic bias and automated discrimination. Corporations like Amazon realize the power of algorithmic accountability. From Richmond to Spokane, Amazon has actively blocked and undermined similar bills for privacy protections so that they can continue to profit off the exploitation of people’s data through their online retail and emerging technologies.¹⁰ Collective people power is greater than consolidated corporate power. That’s why Color Of Change is calling on Congress to pass the Algorithmic Accountability Act.

Join us in advocating for the Algorithmic Accountability Act.

References

1. “Data Capitalism,” Data for Black Lives, https://act.colorofchange.org/go/337317?t=7&akid=52872%2E14915636%2Es2IWY5.
2. Ibid.
3. Tom Feltner and Douglas Heller, High Price of Mandatory Auto Insurance in Predominantly African American Communities, Consumer Federation of America, (November 2015).
4. Kim Lyons, “Facebook’s ad delivery system still has gender bias, new study finds,” The Verge, April 9, 2021, https://act.colorofchange.org/go/337318?t=9&akid=52872%2E14915636%2Es2IWY5.
5. Colin Lecher and Maddy Varner, “NYC’s school algorithms cement segregation. This data shows how,”  May 26, 2021, https://act.colorofchange.org/go/346313?t=11&akid=52872%2E14915636%2Es2IWY5.
6. Jennifer Tsai, “Jordan Crowley would be in line for a kidney—if he were deemed white enough,” Slate, June 27, 2021, https://act.colorofchange.org/go/337319?t=13&akid=52872%2E14915636%2Es2IWY5.
7. Robyn Caplan, Joan Donovan, Lauren Hanson, and Jeanna Matthews, “Algorithmic Accountability: A Primer,” Data & Society, April 18, 2018, https://act.colorofchange.org/go/340900?t=15&akid=52872%2E14915636%2Es2IWY5.
8. “Wyden, Booker and Clarke introduce Algorithmic Accountability Act of 2022 to require new transparency and accountability for automated decision systems,” Ron Wyden United States Senator for Oregon, February 3, 2022, https://act.colorofchange.org/go/346384?t=17&akid=52872%2E14915636%2Es2IWY5
9. Emmanuel Martinez and Lauren Kirchner, “The Secret Bias Hidden in Mortgage-Approval Algorithms,” The Markup, August 25, 2021, https://act.colorofchange.org/go/337316?t=19&akid=52872%2E14915636%2Es2IWY5.
10. Jeffrey Dastin, Chris Kirkham, and Aditya Kalra, “Amazon wages secret war on Americans’ privacy, documents show,” Reuters, November 19, 2021, https://act.colorofchange.org/go/337321?t=21&akid=52872%2E14915636%2Es2IWY5.