Starting Jan. 1, 2027, the unit will enforce the RAISE Act, which requires the largest AI developers to publish safety plans and report serious safety incidents.
New York Attorney General Letitia James created a unit in her office to enforce New York AI laws. The unit will lead enforcement of the Responsible AI Safety and Education Act (RAISE Act). That law sets safety rules for the largest AI developers and takes effect Jan. 1, 2027. James said AI should “never come at the expense of New Yorkers’ safety and privacy.”
The unit is called the AI Safety, Standards, Enforcement, and Transparency Unit. The attorney general’s office is currently building the team.
Frontier AI developers face multimillion-dollar fines
Starting January 1, 2027, the attorney general can seek fines of up to $1 million for a first violation of the RAISE Act and up to $3 million for each subsequent violation. The law applies to companies with $500 million or more in annual revenue that train AI models using more than 10^26 computing operations.
These developers must publish plans for preventing their AI models from causing mass casualties or major damage. They will also have to report serious safety incidents to the state within 72 hours of learning of them. These incidents include a developer losing control of an AI model in a way that causes death or injury. They also include an AI model deceiving its developer to undermine the controls or monitoring the developer put in place.
Developers will register with an oversight office created by the law in the New York Department of Financial Services. They will send their incident reports to that office, and the attorney general will take violations to court.
The unit will support chatbot and AI advertising cases
Beyond the RAISE Act, the unit will lead enforcement of other New York AI laws and advise other divisions of the attorney general’s office on investigations that involve AI. Three of those laws apply to any company that offers a chatbot or uses AI in its ads, not only to AI developers:
- New York’s AI companion law, in force since 2025, requires companies that run AI chatbots to make reasonable efforts to detect and address suicidal thoughts in users, and to tell users they are not talking to a human.
- A law Gov. Kathy Hochul signed in October will bar chatbot companies from offering unsafe features, such as acting human or flattering the user, to anyone they haven’t confirmed is an adult.
- The Synthetic Performer Disclosure Act requires companies to disclose when their ads use AI-generated people in place of human performers.
James sought whistleblowers inside AI companies
The announcement said the unit builds on James’s earlier efforts to protect New Yorkers from AI harms. After reports that AI agents at OpenAI hacked Hugging Face, she encouraged employees at AI companies to file whistleblower complaints with her office. Hugging Face is an AI company that runs a site where developers share AI models.
House bill would bar state safety plan requirements
If Congress passes the American AI Security Act, New York could no longer require the safety plans that the unit will enforce under the RAISE Act. The bill would bar states from requiring AI developers to disclose their safety plans, test results, or risk assessments. Reps. Josh Gottheimer (D-N.J.) and Mike Lawler (R-N.Y.) introduced it in the House on Nov. 10.

