“Anthropic has quietly presented a draft AI safety bill to lawmakers that would restrict the government's ability to block advanced AI model releases. The proposal also seeks to override certain state-level AI safety regulations with federal standards. This marks a significant move by a major AI lab to directly shape the legislative framework governing its own technology.”
Key Takeaways
- Anthropic privately pitched a draft AI safety bill to US lawmakers, according to Politico Tech.
- The proposal would limit government powers to block the release of advanced AI models.
- The bill would preempt certain state-level AI safety laws, favouring a federal framework.
Anthropic is privately lobbying lawmakers with its own draft AI safety bill.
trending_upWhy It Matters
When AI companies author their own safety legislation, they gain enormous influence over the rules that govern them — raising questions about regulatory capture. The preemption of state laws is particularly consequential, as states like California have been the most aggressive in pursuing AI oversight; a federal bill could neutralise those efforts. Competitors and civil society groups will likely scrutinise Anthropic's proposal closely, given the tension between a lab's commercial interests and genuine public safety goals. Watch for how Congress responds and whether other labs, such as OpenAI or Google DeepMind, follow with rival proposals.
FAQ
Why would Anthropic write its own AI safety bill?
By drafting legislation itself, Anthropic can shape the legal boundaries around model releases and safety requirements in ways that align with its operational and commercial interests. It also allows the company to position itself as a responsible actor willing to engage with regulators proactively.
How would this bill affect state AI safety laws?
The draft proposal would preempt certain state-level AI safety regulations, meaning federal rules would take precedence. This is significant because states like California have passed some of the strictest AI oversight measures in the US, and preemption would effectively roll back that authority.
Does this mean the government would have less power to stop dangerous AI releases?
Under the proposal, the circumstances in which the government could block an advanced AI model from being released would be explicitly limited. Critics may argue this weakens public protections, while supporters could contend it provides clearer, more predictable rules for the industry.
