FTC Announces it May Pursue Rulemaking to Combat Discrimination in AI

On December 10, the Federal Trade Commission (FTC) announced it is considering a rulemaking on commercial Artificial Intelligence (AI). The purpose of the rulemaking, according to an advanced notice of proposed rulemaking (ANPRM) titled “Trade Regulation in Commercial Surveillance,” would be “to curb lax security practices, limit privacy abuses, and ensure that algorithmic decision-making does not result in unlawful discrimination.”

While not formally part of the rulemaking process mandated by the Administrative Procedure Act, advanced notices allow agencies to solicit public comment before drafting more specific proposals. The FTC has not yet issued privacy or artificial intelligence rules, though it has indicated that such rulemaking is on the horizon.  The December 10 ANPRM is another signal that the FTC is gearing up to develop substantive privacy guidelines.

The anti-discriminatory focus of the notice reflects concern reflected in a variety of policy papers, research and media coverage over whether commercial AI has the potential to deliver biased outcomes. The movement towards official rulemaking also aligns with the AI fact-finding efforts and interests that the FTC has demonstrated at its last two PrivacyCon events, including a specific focus on advertising, fairness, and transparency in AI and algorithms at the summer 2020 PrivacyCon.

The language of the ANPRM is broad, and could cover a variety of commercial practices. Earlier statements, however, may provide insight into the FTC’s thought-process. The Commission has previously pointed to protected-class bias in healthcare delivery and consumer credit as prime examples of algorithmic discrimination. See Jillson, E., Aiming for truth, fairness, and equity in your company’s use of AI, FTC Business Blog (April 19, 2021). Additionally, in a recent semiannual Statement of Regulatory Priorities,  the FTC expressed that, among the many pressing issues confronted by consumers in the modern economy, “abuses stemming from surveillance-based business models are particularly alarming.” Therefore, the FTC may also target algorithms that drive behavioral ad-based ecosystems.

As we have highlighted in our prior posts, the FTC recently updated its rulemaking procedures. The ANPRM contemplates rulemaking under section 18 of the FTC Act, which authorizes the FTC to promulgate “rules which define with specificity acts or practices which are unfair or deceptive acts or practices in or affecting commerce.” 15 U.S.C. Sec. 57a. The FTC’s process for issuing section 18 rules, also known as “Trade Regulation Rules,” or “Magnuson-Moss Rules,” has recently been streamlined in order to facilitate the FTC’s renewed efforts to create substantive guidelines for the modern economy.

While the FTC must still take several steps before any rules become final, categorizing certain uses of commercial AI as unfair or deceptive under the FTC Act could be a further step towards the creation of uniform federal privacy standards.

Interested parties should expect the comment period under either the ANPRM or a more formal rulemaking to begin in February 2022.

, , ,