This voice is automatically generated. Please let us know if you have any feedback.
Diving overview:
- New York Governor Kathy Hochul signed bill According to , it is establishing a framework for safety and transparency for Frontier AI models. Friday announcement.
- of The Responsible AI Safety and Education (RAISE) Act Large AI developers will need to create and publish information about their safety protocols. It also requires developers to report incidents related to model safety to the state. 72 hours According to the statement, they learned that the incident had occurred. law comes into force January 1, 2027.
- “By enacting the RAISE Act, New York State is once again leading the nation in setting strong and sensible standards for frontier AI safety and holding major developers accountable to safety and transparency protocols.” Hochul said in the announcement.
Dive Insight:
Hocheol”The move to sign safety and transparency requirements for AI models into law is a blow; President Donald TrumpExecutive order calls into question the ability of US states to regulate technology.
Hochul explained raise act as a “state-led approach to AI safety” inspired by other states that regulate this technology. Includes California's Frontier Artificial Intelligence Transparency Act.
“This legislation builds on California's recently adopted framework, which creates uniform standards among the nation's leading technology states at a time when the federal government has lagged behind and failed to put in place common-sense regulations that protect the public.” Hochul Said.
This is the first state AI law enacted since President Trump signed the executive order. presidential order Earlier this month, AI Litigation Task Force Challenging state AI laws that interfere with existing federal law or “unconstitutionally regulate interstate commerce.” The order is the culmination of a year spent removing enforcement powers from federal agencies; reduced regulation. This also coincides with protests from technology companies. A messy patchwork of state AI laws.
President Trump's AI Litigation Task Force to be established in the US 30 days of his executive order. It remains to be seen which states will be targeted first, but Colorado is specifically singled out in the executive order for its problematic AI laws. Colorado AI Law; Prohibit algorithmic discrimination.
New York State's RAISE Law Large-scale AI model developers, such as OpenAI and Anthropic, must document technical and organizational safety and security protocols, as well as testing and evaluation procedures. Developers are also required to appoint a senior human resources member responsible for ensuring compliance.
The law introduces a 72-hour reporting requirement for AI model safety incidents. This may include the Frontier Model autonomously performing actions other than those requested by the user, significant failure of technical controls, theft, and malicious use or unauthorized access to the Frontier Model.
Developers who fail to file required RAISE Act reports or make false statements may be subject to civil penalties of up to $1 million for a first violation and up to $3 million for subsequent violations.
“Big tech oligarchs think it's okay to put their profits ahead of our safety, and we disagree,” state Sen. Andrew Gounardes of New York said in a statement Friday. “This legislation makes clear that innovation and safety do not have to be at odds.”
