Major artificial intelligence developers are locking horns in Massachusetts over proposed state regulations that could become the most stringent in the nation. The state Senate recently passed an economic development bill requiring leading AI labs to submit to independent evaluations of catastrophic risks every 120 days. These external evaluations would assess whether frontier models pose potential threats that could lead to mass casualties or major structural property damage.

The legislation has split top industry players, with Anthropic backing the push for stricter state-level oversight while competitors argue the measure goes too far. Anthropic has actively supported the proposal, calling it “the clearest and strongest AI legislation in the country.” Company leaders maintain that third-party reviews are essential, noting that “we ultimately don’t think the industry should grade its own homework.”

Conversely, OpenAI opposes the strict auditing timeline, warning that frequent assessments could severely delay the release of critical cybersecurity defenses. Representatives from OpenAI prefer a national baseline or a setup matching Illinois laws, emphasizing that inconsistency “doesn’t mean safer. It just means confusion.”

The debate puts state lawmakers in a delicate balancing act, as Governor Maura Healey attempts to attract tech investments while building up the local economy. State leaders are currently debating the final details of the measure, trying to establish clear safety boundaries without driving away technical innovation and top-tier startup talent.