The Commerce Department’s National Telecommunications and Information Administration (NTIA) announced a request for comment on accountability measures for artificial intelligence tools amid growing concerns about the safety of generative AI following the release of ChatGPT-4. The comments collected by the NTIA will inform the policy guidance provided to U.S. policy makers. While the NTIA serves as an advisor rather than a regulator, its advice will likely be used as a pathway to federal regulation.
Audit standards already in the making: As previously reported, employers and AI vendors are already scrambling to establish standards ahead of the effective date of a recent New York City law which requires notice and audit requirements for AI decision-making tools and places responsibility for such audits on employers. California is in the process of developing AI regulations covering “developers” and “deployers” (e.g., employers). The European Union is currently debating the AI Act, a broad piece of legislation aimed at setting strict limits on AI tools, and China has said it will require AI audits beginning in May.
Federal regulation lags: Despite the emergence of legislation across the U.S. and globally, AI is largely unregulated throughout the world. Recognizing the lack of unified standards, industry leaders have called for a six-month moratorium on the development of AI, arguing that “powerful AI systems should be developed only once we are confident that their effects will be positive and their risks will be manageable.”
Schumer floats regulatory framework. According to Axios, Senate Majority Leader Charles Schumer (D-NY) has been quietly developing a framework that could lead to legislation focused on four pillars:
- Identification of who trained the algorithm and its intended audience.
- Disclosure of its data source.
- An explanation for how it arrives at its responses.
- Transparent and strong ethical boundaries.
Outlook: Public comments on the Commerce Department request will be open for 60 days after which the NTIA will produce a report with specific policy recommendations. The agency asked for consideration on whether rules should differ depending on the sensitivity of the area, such as when AI is used for decisions related to employment or health care. HR Policy will be submitting comments based on the previous input of its AI working group.