New York’s RAISE Act forces OpenAI and Anthropic to register and report safety incidents
What happened
The state of New York has opened a registration portal for large frontier artificial‑intelligence developers, a step that begins enforcement of the RAISE Act on January 1 2027. Companies that generate more than $500 million in annual revenue, such as OpenAI and Anthropic, must complete the registration by that date. The portal is managed by the newly created Office of Digital Innovation, Governance, Integrity, and Trust (DIGIT).
Under the RAISE Act, the listed developers are required to publish detailed safety frameworks on their public websites, issue model‑transparency reports, and report any critical safety incident within 72 hours of occurrence. Violations can attract fines of up to $1 million for a first offense and $3 million for subsequent offenses, and the state may also pursue enforcement actions against firms that fail to register. Governor Kathy Hochul announced the law alongside the launch of the New York Center for AI Responsibility and Research at SUNY Binghamton, which will independently assess AI safety and work with DIGIT on compliance.
The RAISE Act joins other state‑level AI regulations, including California’s SB 53 and Illinois’ SB 315, which impose similar safety‑framework publishing and incident‑reporting duties. New York also enforces a companion‑bot disclosure law and a forthcoming chatbot‑safety law for minors that will take effect in June 2027. These measures collectively aim to create a national model for AI oversight that prioritises community safety over unchecked technological deployment.
Why it matters
For businesses in Rwanda and the wider East African region, the New York law illustrates how governments are beginning to impose concrete accountability standards on powerful AI providers. While the regulation is U.S.‑focused, many of the same AI models are offered globally through cloud services. Companies that rely on these services may soon need to demonstrate compliance with safety‑framework disclosures and rapid incident reporting, especially if they serve customers in jurisdictions that adopt similar rules.
Rwandan startups and developers can use this development as a prompt to adopt proactive safety practices now, such as documenting model limitations, establishing internal incident‑response procedures, and preparing transparency reports. Early alignment with emerging international standards could reduce future regulatory friction, build trust with clients, and position local firms as responsible AI innovators in a market that increasingly values ethical deployment.
Key takeaways
- New York’s RAISE Act requires large AI labs to register, publish safety frameworks and report incidents within 72 hours.
- Penalties for non‑compliance can reach $3 million, and the state has created a dedicated office to enforce the rules.
- Rwandan businesses using frontier AI should consider adopting similar safety documentation and reporting practices.
AI-assisted summary published by the Diolichat Team. See our editorial policy. Spotted a mistake? Email info@diolichat.co.rw.