California Senator Scott Wiener Introduces Amendments to SB 53, Mandating AI Safety Reports for Top Developers.

California State Senator Scott Wiener has reintroduced amendments to his bill SB 53, requiring top AI companies to publish safety and security protocols and issue reports on incidents. If passed, California would be the first state to impose transparency requirements on leading AI developers, including OpenAI, Google, and Anthropic. The bill aims to strike a balance between transparency and the growth of California's AI industry.

California State Senator Scott Wiener has reintroduced amendments to his bill, SB 53, which aims to mandate transparency requirements for the world's leading AI companies. If passed, California would become the first state to impose such regulations, impacting major players such as OpenAI, Google, and Anthropic [1].

The bill, initially introduced as SB 53, seeks to balance transparency with the growth of California's AI industry. It requires companies to publish safety and security protocols and issue reports when safety incidents occur. The amendments are heavily influenced by recommendations from a policy group formed by Governor Gavin Newsom, which emphasized the need for industry to publish information about their systems to create a robust and transparent evidence environment [1].

The bill also introduces whistleblower protections for employees who believe their company's technology poses a critical risk to society, defined as contributing to the death or injury of more than 100 people, or more than $1 billion in damage. Additionally, it proposes the creation of CalCompute, a public cloud computing cluster to support startups and researchers developing large-scale AI [1].

SB 53 is currently headed to the California State Assembly Committee on Privacy and Consumer Protection for approval. If it passes, the bill will need to navigate several more legislative bodies before reaching Governor Newsom's desk. Meanwhile, New York Governor Kathy Hochul is considering a similar AI safety bill, the RAISE Act, which would also require large AI developers to publish safety and security reports [1].

The proposal comes as federal lawmakers considered a 10-year AI moratorium on state AI regulation, aiming to prevent a patchwork of laws. However, this proposal failed in a 99-1 Senate vote earlier in July, allowing states to continue their efforts [1].

The bill has faced resistance from some AI companies, with OpenAI, Google, and Meta being more resistant to transparency requirements. However, Anthropic has endorsed the need for increased transparency. Leading AI model developers typically publish safety reports, but their consistency has waned in recent months, with companies like Google and OpenAI not publishing reports for their most advanced models [1].

SB 53 represents a toned-down version of previous AI safety bills but could still force AI companies to publish more information than they currently do. Companies will be closely watching as Senator Wiener tests these boundaries once again [1].

References:
[1] https://techcrunch.com/2025/07/09/california-lawmaker-behind-sb-1047-reignites-push-for-mandated-ai-safety-reports/
[2] https://news.bloomberglaw.com/states-of-play/california-lawmaker-pushes-ai-firms-to-release-safety-policies