U.S. Senator Rob Portman (R-OH), Ranking Member of the Senate Homeland Security and Governmental Affairs Committee and Co-Chair of the Senate Artificial Intelligence (AI) Caucus, introduced the Stopping Unlawful Negative Machine Impacts through National Evaluation Act to clarify that existing civil rights laws apply to decisions made by AI systems just as if those decisions were made by humans. The Stopping Unlawful Negative Machine Impacts through National Evaluation Act would make it clear that existing civil rights laws apply to both AI-enabled and human decisions alike, and authorizes the National Institute of Standards and Technology (NIST) to set up “technology evaluations” to help industry find innovations which reduce algorithmic discrimination. 

“Amidst its great promise, artificial intelligence does have pitfalls,” said Senator Portman, Co-Chair of the Senate AI Caucus. “We know that powerful algorithms can discriminate against individuals, which is why it’s vital that Congress takes steps to ensure that AI systems are safe and robust. Rather than jump all the way to heavy-handed regulation of this new technology, Congress should make clear that we already have numerous laws on the books to guard against discrimination on the basis of protected characteristics. By clarifying that our civil rights laws apply to all decisions whether made by human or machine, we can take a valuable first step to ensure that AI systems work for people, and not against them.” 

Tech evaluations are a collaborative, non-regulatory processes where NIST brings in technologists, companies, and scientists to work quickly and in a focused way on a very specific tech problem. Rather than regulating AI, this legislation would reduce bias by expanding the tech evaluation model to different high-risk uses of AI to empower industry and government to work together to innovate and solve problems in a collaborative rather than adversarial way. 

Original source can be found here