US proposes Algorithmic Accountability Act to stamp out bias in tech firms

US lawmakers proposed a new bill on Wednesday that would require large tech firms to remove any algorithm bias embedded in their computer systems.

The bill, entitled the Algorithmic Accountability Act of 2019, would authorise the Federal Trade Commission to create regulations requiring companies to assess their systems for race, gender, or other discriminatory biases.

The proposed legislation would apply to companies with annual revenue of above $50m (£38m) or data brokers and businesses that hold data on more than one million people.

Algorithm bias

“Computers are increasingly involved in the most important decisions affecting Americans’ lives – whether or not someone can buy a home, get a job or even go to jail,” Democratic Senator, Ron Wyden, said in a press release announcing the bill.

“But instead of eliminating bias, too often these algorithms depend on biased assumptions or data that can actually reinforce discrimination against women and people of color.”

The proposal comes as major tech companies continue to fight against accusations of racial, gender and political bias.

Facebook charge

The announcement cited a recent charge from the Department of Housing and Urban Development that accused Facebook with violating the Fair Housing Act by allowing advertisers to restrict who saw the ads for homes on the basis of their race, religion and disability status.

The bill, which is also backed by Democratic Senator, Cory Booker, and representative, Yvette Clarke, is endorsed by tech civil rights groups, including Data for Black Lives, the Centre on Privacy and Technology at Georgetown Law and the National Hispanic Media Coalition.

“Automated decisions are not neutral decisions. They turn on human data. As long as humans are biased, algorithms will be biased too, the Centre on Privacy and Technology at Georgetown said in a statement. This bill will force companies to reckon with that reality.”