Disagreements reemerge in bipartisan AI safety talks - Semafor

Disagreements over how far to go on AI safety legislation portend a rocky road ahead for bipartisan Senate talks on a bill that remains the most likely vehicle for action this year.

The Senate Commerce Committee’s lead Democrat, Maria Cantwell of Washington, brought her concerns to X on Wednesday, saying that while she “welcomes the urgency” to get something done, the “answer is not a weak federal standard that becomes a backdoor for wiping out stronger state protections.”

At issue: A bill in the works from Commerce Chair Ted Cruz, R-Texas, Sen. Amy Klobuchar, D-Minn., and Majority Leader John Thune, which is said to still contain language preempting state-level AI safety laws.

A source familiar with the matter told Semafor that Cantwell still wants a stricter version of the bill, which creates legal duties for AI companies to manage catastrophic risks.

The source said Cantwell, Anthropic, and safety groups are “refusing to play ball” and unwilling to create a bipartisan product to have federal power over AI models “while ensuring continued innovation and due process in the courts.”

A source within Anthropic said the company hasn’t taken a position on the bill and continues to provide feedback to the bill sponsors.