New York Signs Nation-Leading AI Safety Law as Federal Policy Lags
Summary
On December 19, 2025, New York Governor Kathy Hochul signed the Responsible AI Safety and Education (RAISE) Act into law. The bill requires large AI developers to publish safety frameworks, report critical incidents, and establish transparency protocols. It also creates an oversight office to monitor compliance. The law builds on similar efforts in California, setting a de facto standard in the absence of federal regulation. The move underscores growing efforts by states to shape how artificial intelligence is developed and deployed.
What It Means
The RAISE Act represents one of the most comprehensive state-level AI safety frameworks in the United States. By requiring developers to disclose safety practices and report incidents, the law promotes greater transparency around advanced AI models. This could influence how companies document and share information about potential risks, errors, or harmful outcomes.
For AI companies, the law introduces new compliance responsibilities soon to take effect. Large developers will need to align their internal safety processes with the reporting standards set by the state. In some cases, this could affect product release timelines or internal practices for managing model behaviour.
From an industry perspective, New York and California together are shaping a patchwork of state guidance that may act as a de facto national benchmark in the absence of a unified federal framework. This could lead other states to adopt similar standards, encouraging broader industry adoption of safety and transparency measures even for companies based outside those states
Key Takeaways
New York enacted the RAISE Act to regulate advanced AI model safety. Large AI developers must create and publish safety frameworks.
Critical incidents must be reported to state regulators.
The law establishes a new oversight office to enforce transparency.
State policy may influence broader U.S. AI governance in the absence of federal law.
Our Take (2026 Outlook) * Speculative
As 2026 unfolds, we may see more states develop comparable AI safety laws, particularly if industry and public interest groups support standardized disclosure and transparency. Companies that proactively adapt to these frameworks could gain trust and stability as AI governance evolves.
References
Axios, 2025
Bloomberg Law, 2025
Governor’s Press Office (New York), 2025
CryptxAI publishes simplified AI and crypto downloadable briefings.
Comments