Recent legislative developments in New York have established the state as a leader in Artificial Intelligence (AI) regulation with the enactment of the “Responsible AI Safety and Education Act” (RAISE Act). On December 19, 2025, Governor Kathy Hochul signed the legislation (S6953B/A6453B), which she described as setting a “nation-leading standard” for managing the safety and transparency of advanced AI systems.
Background: Managing “Frontier” AI Risks
As AI technology evolves at an unprecedented pace, concerns regarding public safety and critical infrastructure have grown. The RAISE Act specifically targets “frontier models” which are advanced AI systems that require significant computational power and exceed $100 million in training costs to develop.
The legislation aims to mitigate risks related to “critical harm,” which includes scenarios involving mass casualties or substantial economic damages. Governor Hochul emphasized that while AI drives groundbreaking advances, the state must also hold developers accountable for safety protocols to protect the public.
Key Provisions Introduced by the Act
The RAISE Act imposes strict compliance obligations on “large developers” of these high-threshold AI models. Key components of the law include:
- Safety Protocols. Developers must implement and maintain written safety and security protocols designed to prevent their models from causing critical harm.
- Reporting Requirements. Large developers are required to report any “safety incidents” to the New York Attorney General within 72 hours of discovery.
- Oversight Mechanism. The Act establishes a new oversight office within the Department of Financial Services to assess developer compliance and publish annual reports.
- Transparency. Developers must publish redacted versions of their safety protocols, ensuring a level of public transparency while protecting proprietary information.
- Penalties. The Attorney General can bring civil actions for violations, with penalties up to $1 million for a first violation and up to $3 million for subsequent violations.
Next Steps for Employers
While the RAISE Act primarily targets the developers of massive AI models (such as those creating foundational Large Language Models), the downstream effects will impact employers who utilize these tools. To prepare for this new regulatory environment, employers should:
- Assess AI Utilization: Conduct an audit of the AI tools and systems currently in use within your organization to identify if any rely on “frontier models” regulated by this Act.
- Review Vendor Relationships: Engage with third-party software vendors to ensure they are aware of and compliant with the new transparency and safety obligations.
- Prepare Governance Frameworks: Begin developing or updating internal AI usage policies that address safety, data privacy, and risk mitigation to align with the state’s focus on responsible AI.
- Monitor Regulatory Updates: As the new oversight office is established, stay informed on future guidance that may clarify how these regulations affect end-users and corporate compliance strategies.
