The political tipsheet Axios reports this week that Senate Majority Leader Chuck Schumer (D-NY) has taken the first steps toward legislation to regulate the use of artificial intelligence, circulating a general framework for new rules among academic, industry and government AI experts. The U.S. Commerce Department, the Copyright Office, the Patent and Trademark Office and other government agencies have taken preliminary steps toward possible regulatory action on AI within their respective domains, but the recent movement in the Senate is the first indication that broader legislative action could be in the works.
By spearheading the effort himself, Schumer is also signaling a degree of urgency behind the effort. Rather than working through the normal committee process, which can be time-consuming and uncertain, particularly when multiple committees would likely have jurisdiction over various parts of any bill, Schumer may be looking to use his power as majority leader to control the Senate schedule to bring a bill directly to the floor.
According to the reports, the framework outlined by Schumer is aimed at transparency in AI models and focused on four key areas:
- Identifying who trained the model and who the intended audience is;
- Disclosing the source or sources of the training data;
- Describing and explaining how a generative AI model arrives at its responses;
- Strong ethical boundaries.
Any bill that comes out of the process would require 60 Senate votes to overcome a Republican filibuster. Concern over AI has come from both sides of the aisle, however, and by keeping the measure focused on transparency and disclosure Schumer may be able to muster the necessary majority. (Update: On Tuesday, several Republican senators told Fox News they were open to discussing a possible framework for AI regulation).
It is also likely to draw support from the White House. In October, the White House Office of Science and Technology Policy issued a Blueprint for an AI Bill of Rights that also focused on transparency, and in January a White House task force issued a report and roadmap for developing a national research infrastructure for AI aimed at “broaden[ing] access to the resources essential to artificial intelligence (AI) research and development.”
So far, there are not indications of movement on AI issues in the House of Representatives.