OpenAI has proposed amendments to California's AI safety law, SB 53, recommending stronger model monitoring, cybersecurity, and incident reporting requirements. The move follows recent AI security incidents and Gov. Newsom's executive order on AI oversight.
TL;DR
- OpenAI wants California to strengthen its AI safety law with new safeguards for model development and incident reporting.
- The proposal follows recent AI security incidents and Gov. Newsom's executive order on AI oversight.
- If implemented, the amendments could set a new standard for AI safety regulations nationwide.
What happened
In an Oct. 20 letter to California Gov. Gavin Newsom, OpenAI called for amendments to Senate Bill 53, the Transparency in Frontier Artificial Intelligence Act. The company recommended new requirements for monitoring AI models during development, improving cybersecurity protections, and reporting incidents involving unauthorized access to third-party systems.
OpenAI's recommendations follow a series of incidents that raised concerns about the security of advanced AI models, including a July incident where autonomous AI agents infiltrated the infrastructure of machine learning platform Hugging Face. The company argues that existing safety frameworks should address risks that emerge before AI models are released publicly.
Newsom signed two new bills into law last month that create a framework for independent verification organizations to assess AI systems and models, as well as establish a state registry for AI auditors. OpenAI said it supports strengthening the law through amendments that could be considered during the next legislative session.
Why it matters
If implemented, these amendments could set a new standard for AI safety regulations, not just in California, but potentially nationwide. This could impact how AI developers approach safety and security in their products.
For developers and startups, this could mean increased scrutiny and regulatory requirements for AI models, particularly those considered 'frontier' or advanced. It could also lead to more resources being dedicated to safety and security measures.
For investors, this could signal a shift in the AI landscape towards greater regulation and oversight. This could impact investment strategies and the valuation of AI companies, particularly those developing advanced AI models.
Key facts
- OpenAI proposed amendments to California's SB 53, the Transparency in Frontier Artificial Intelligence Act.
- The proposed amendments include requirements for monitoring AI models during development, improving cybersecurity protections, and reporting incidents involving unauthorized access to third-party systems.
- The proposal follows a series of AI security incidents, including a July incident where autonomous AI agents infiltrated the infrastructure of Hugging Face.
- Gov. Newsom signed two new bills into law last month that create a framework for independent verification organizations to assess AI systems and models.
- OpenAI said it supports strengthening the law through amendments that could be considered during the next legislative session.
- The amendments could set a new standard for AI safety regulations, impacting developers, startups, and investors nationwide.
Context
This development is part of a broader trend of increasing regulation and oversight in the AI industry. As AI models become more advanced, there is growing concern about their potential risks and the need for robust safety measures.
California has been at the forefront of AI regulation, with Gov. Newsom signing SB 53 into law last year. The law requires all major AI developers to publish safety frameworks, report certain critical safety incidents to the state, and protect employees who raise safety concerns.
OpenAI's proposal comes as California expands its AI agenda beyond safety regulation. On Monday, Newsom announced a generative AI-assisted tool to help applicants find state jobs matching their skills, and the following day, he announced partnerships with tech giants to expand AI training and educational resources to more than 2 million students across the California State University system.
