New York State Passes Landmark AI Safety Bill
In a significant move on Thursday, New York state lawmakers successfully passed a groundbreaking bill aimed at regulating the use of frontier AI models developed by tech giants such as OpenAI, Google, and Anthropic. The bill, known as the RAISE Act, seeks to prevent these advanced AI systems from causing catastrophic scenarios that could result in the loss of more than 100 lives or over $1 billion in damages.
The passage of the RAISE Act marks a pivotal moment for the AI safety movement, which has faced challenges in recent years due to the prioritization of speed and innovation over safety concerns. Notable figures in the field of AI, including Nobel laureate Geoffrey Hinton and research pioneer Yoshua Bengio, have been strong advocates for the bill. If enacted into law, the RAISE Act would establish the first set of legally mandated transparency standards for frontier AI labs in the United States.
Drawing parallels to California’s SB 1047, which was vetoed, the RAISE Act incorporates similar provisions but aims to strike a balance that does not stifle innovation among startups and academic researchers, as criticized in the former bill. Senator Andrew Gounardes, a co-sponsor of the RAISE Act, emphasized the urgency of implementing safeguards in light of the rapid evolution of AI technology and the inherent risks involved.
The bill now awaits approval from New York Governor Kathy Hochul, who will have the option to sign it into law, request amendments, or veto it entirely. Should it be enacted, the RAISE Act would mandate the world’s leading AI labs to release comprehensive safety and security reports on their frontier AI models, as well as report any safety incidents that may arise. Failure to comply with these standards could result in civil penalties of up to $30 million enforced by New York’s Attorney General.
The RAISE Act specifically targets major tech companies, regardless of their location, and sets transparency requirements for AI models trained using substantial computing resources and made available to New York residents. Despite pushback from Silicon Valley, Assemblymember Alex Bores assured that the regulatory burden imposed by the bill is manageable and should not deter tech companies from operating in New York.
While some concerns have been raised about the potential impact on smaller companies, Senator Gounardes clarified that the bill is not intended to apply to such entities. As discussions continue, it remains to be seen how tech giants like OpenAI, Google, and Meta will respond to the RAISE Act and its implications for their AI models.
In conclusion, the passage of the RAISE Act underscores the growing importance of AI safety regulations in an era of rapidly advancing technology. By setting a precedent for transparency and accountability in the AI industry, New York state is taking a proactive stance towards mitigating potential risks associated with powerful AI systems.
