Massachusetts Senate passes bill requiring 120-day AI risk reviews
This digest was compiled by AI from multiple sources — links to the originals are below.
The Massachusetts Senate passed an economic development bill last month requiring leading AI labs to undergo independent reviews of catastrophic risks from their frontier models at least once every 120 days, a first for a US state. Review findings would be made public, but the state could not use them to halt AI development. Anthropic calls the proposal 'the clearest and strongest AI legislation in the country,' while OpenAI warns the reviews would slow release of cybersecurity models.
Key Facts
- The Massachusetts Senate bill requires leading AI labs to undergo independent reviews of catastrophic risks from their frontier models at least once every 120 days, a first for a US state.
- Review findings would be made public, but the state could not use them to halt AI development.
- Anthropic calls the Massachusetts proposal 'the clearest and strongest AI legislation in the country,' while OpenAI warns the reviews will slow release of cybersecurity models and favors a uniform Illinois-style standard.
- The Massachusetts proposal empowers outside organizations to evaluate model dangers every few months, assessing potential catastrophes that kill or seriously injure at least 50 people or cause $1 billion or more in property damage.
Senate Safeguards
The Massachusetts Senate passed an economic development bill last month that includes a first-in-the-nation requirement for leading AI labs to undergo independent reviews of catastrophic risks from their frontier models at least once every 120 days. Review findings would be made public, though the state could not use them to halt AI development. The provisions go further than recent laws in California, New York and Illinois. Independent evaluators would assess the potential for catastrophes that kill or seriously injure at least 50 people or lead to $1 billion or more in property damage.
Industry Divide
Anthropic PBC has advocated for stricter state regulation in the absence of a federal standard and calls the Massachusetts proposal 'the clearest and strongest AI legislation in the country.' OpenAI warns the reviews will slow the release of cybersecurity models that might protect against the risks lawmakers fear. Donnie Fowler, OpenAI's head of US state policy and partnerships, said inconsistency 'doesn't mean safer — it just means confusion.' OpenAI prefers states adopt a uniform standard in line with the Illinois law, which requires a third-party audit once a year. The Massachusetts proposal includes a similar annual compliance review but separately empowers outside organizations to make evaluations of a model's dangers, independent of the developer's guidelines, every few months.
Legislative Road Ahead
Governor Maura Healey, a Democrat running for reelection in November, did not include the AI safeguards passed by the Senate in her initial economic development proposal. Both she and House lawmakers still must agree to the provision. The governor's office is concurrently seeking to build up the region's AI sector and curb the outflow of startup talent.
1 source
Massachusetts Senate passes bill requiring 120-day AI risk reviews
