Situation assessment: The biggest AI labs in the country now face a hard deadline from a second major state regulator. New York opened a registration portal on Tuesday, and large frontier AI developers, including OpenAI and Anthropic, must register by January 1, 2027. TechCrunch AI reports that this kicks off enforcement of the RAISE Act, one of the most significant AI safety laws in the U.S.
For AI labs, this is a compliance problem. For everyone watching how AI gets governed, it’s a sign that state rules are starting to work as a group.
🎯 What the RAISE Act Requires
The law applies to “large frontier developers,” meaning companies with more than $500 million in annual revenue. Starting January 1, those companies must:
- Publish safety frameworks on their websites.
- Report critical safety incidents within 72 hours.
- Release public transparency reports for their models.
- Register with the state through the new portal.
The fines are meant to sting. A first violation can cost up to $1 million, and later violations up to $3 million. According to TechCrunch AI, the state can also go after companies that don’t register, though a spokesperson declined to say exactly how.
🏛️ Who’s Enforcing It
New York built new machinery for this law. The Office of Digital Innovation, Governance, Integrity, and Trust (DIGIT) will implement the RAISE Act and oversee the big labs.
It won’t do that alone. On Tuesday, Governor Kathy Hochul also launched the New York Center for AI Responsibility and Research at SUNY’s Binghamton University. The center will independently evaluate the safety, security and operation of advanced AI systems. It’ll also work with DIGIT to review the safety frameworks companies submit. Hochul first announced the center in January, backed by $55 million in donations and capital.
That pairing is what stands out to me. Most state AI laws rely on companies grading their own homework. New York is setting up an independent technical body to check the work, which makes the published frameworks much harder to treat as PR documents.
🗣️ The Political Signal
Hochul didn’t soften her message.
“Tech moguls tell us that their vision for AI is inevitable,” she said during the press briefing. “They tell us get on board or get left behind… That any collateral damage is worth it as long as we win the race to the AI cliff.”
She called the approach “a national model for smart, sensible AI regulation, a model that puts communities above corporations and people above powerful technology.”
That tone matters. It tells you New York isn’t planning to go easy on enforcement.
🗺️ The State-by-State Map
New York isn’t acting alone. Here’s how the major frontier AI laws line up:
- California SB 53 (in effect since January 2026): published safety frameworks, whistleblower protections, critical incident reporting within 15 days.
- New York RAISE Act (effective January 1, 2027): published frameworks, 72-hour incident reporting, transparency reports, mandatory registration.
- Illinois SB 315 (effective early 2027): requirements similar to California’s, plus independent safety compliance audits.
Notice the trend. Each new law goes a bit further. California set the 15-day reporting window, and New York cut it to 72 hours. Illinois adds outside audits. Labs that built their compliance programs around California will need to tighten them.
New York also has a broader AI safety record. Last year it started enforcing a law that requires AI companion apps to remind users they’re talking to a bot and to make reasonable efforts to detect suicidal ideation and self-harm. In October, Hochul signed a chatbot safety law for minors that restricts deceptive human-like behavior, manipulative or sycophantic responses, and encouragement of suicide or disordered eating. That one takes effect June 1, 2027.
⚡ Immediate Implications
- Labs have about seven weeks to register and get their frameworks published.
- The 72-hour window is now the strictest standard. Labs will likely build incident response around it nationwide rather than juggle different state deadlines.
- Smaller players aren’t covered yet. The $500 million revenue threshold leaves startups out, but fast-growing companies should watch that line closely.
- Expect more states to follow. Three big states with overlapping rules make it more likely others copy them, and that adds pressure for a federal standard.
🔭 Outlook
The real test starts in January. Watch for the first transparency reports, the first incident disclosures, and whether DIGIT actually enforces the registration rule. If New York’s independent review model works, other states will probably copy it. Full details are available in the original TechCrunch AI report.