The Algorithmic Accountability Act of 2022 would require companies that use algorithms in select business operations to report on how they work and what data they use.
Private sector companies regularly use algorithms to determine whether to loan consumers money, rent them a home or provide them with legal services. But some systems can contain hidden bias that discriminates against people of a certain race, age, gender or background.
A team of Democratic lawmakers from both chambers introduced new legislation that targets how the algorithms behind everyday software devices and applications work, emphasizing equity and transparency.
Sens. Ron Wyden, D-Ore., and Cory Booker, D-N.J., along with Rep. Yvette Clarke, D-N.Y., introduced the Algorithmic Accountability Act of 2022 on Thursday. It follows a bill of the same name introduced back in 2019, featuring several updates on the technical details the bill intends to regulate.
“If someone decides not to rent you a house because of the color of your skin, that’s flat-out illegal discrimination,” Wyden said in a press release. “Using a flawed algorithm or software that results in discrimination and bias is just as bad. Our bill will pull back the curtain on the secret algorithms that can decide whether Americans get to see a doctor, rent a house or get into a school.”
One imperative the bill hopes to instill is requiring tech companies to conduct impact assessments on the automated systems and algorithms it uses in its business decision making. Some of the issues these reports will look for include hidden biases and discriminatory practices rooted in these algorithms..
The Federal Trade Commission would work alongside lawmakers to execute provisions outlined in the bill and be tasked with creating a guidance framework for these review assessments. Following the completion of these reports, the FTC would also be required to publish aggregated trends seen in algorithm and automation functions, in addition to relevant data sources and metrics.
This information summary would then be made available to the public.
“Discrimination and bias can’t be left unchecked just because the decision is being made by an automated system and a faulty algorithm,” Sen. Mazie Hirono, D-Hawaii, one of the bill’s cosponsors. “This bill will require companies to look at the impact of their automation and provider consumers with the knowledge of when and how they’ve been impacted. Consumers deserve fair and equitable treatment.”
Per the bill’s stipulations, companies subjected to these rules would earn over $50 million in annual gross revenue or over $250 million in equity value over the course of a three year period. They also use software to make decisions surrounding education, employment, family planning, financial services, housing, healthcare or legal services.
Should the bill become law, the FTC will receive funding to hire 50 staff members as part of a new Bureau of Technology to oversee the impact assessments and subsequent summaries.
The Algorithmic Accountability Act marks the latest legislative component aimed at regulating Big Tech companies, along with the recently reintroduced EARN IT Act.
“With our renewed Algorithmic Accountability Act, large companies will no longer be able to turn a blind eye towards the deleterious impact of their automated systems, intended or not,” Clarke said. “We must ensure that our 21st century technologies become tools of empowerment, rather than marginalization and seclusion.”