U.S. Senator calls on regulators to consider bias in facial recognition and AI

American Senator Kamala Harris (D-CA) has sent letters to the FBI, the Federal Trade Commission (FTC), and the Equal Employment Opportunity Commission (EEOC), urging the agencies to consider the risk of bias in artificial intelligence systems, and develop guidelines to ensure fairness in the technology’s application, TechCrunch reports.

The letters, each of which is co-signed by other senators, each pose a set of questions to the receiving agency.

The letter to the EEOC expresses concern with the possibility of AI-based recruiting tools assessing candidates based on their similarities to company leadership, which could entrench inequality with a false veneer of scientific objectivity. The FTC is warned of the possibility of “good faith users” implementing algorithms with biases built into them, and unintentionally engaging in discrimination. It also asks if failing to disclose system bias to customers violates the FTC Act.

“Consider, for example, a situation in which an African American female in a retail store is misidentified as a shoplifter by a biased facial recognition technology and is falsely arrested based on this information,” Harris writes to the FTC. “Such a false arrest can cause trauma and substantially injure her future house, employment, credit, and other opportunities.”

The letter to the FBI refers to a report from the Government Accountability Office on facial recognition which made specific recommendations to it more than two years ago, asking if they have been implemented.

Representatives for Senator Harris told TechCrunch that the move is intended to advance the government’s position, and rouse regulatory agencies to signal companies that they are paying attention to the issue.

Other efforts by members of congress to engage the issue include a call for an investigation by the Department of Justice and letter to individual companies.

“What we’re interested in is the fairness outcome rather than one particular company’s practices,” the representative said.

While some technology companies have been publicly working to combat algorithmic bias, a recent survey indicated that most Americans believe technology does more to entrench bias than combat it.

Article Topics

artificial intelligence  |  biometric-bias  |  biometrics  |  facial recognition  |  false arrest  |  United States