Anthropic's proposal is quite solid. AI is advancing too quickly, and policies definitely need to keep up. Looking forward to seeing how different countries implement it.

CoinWorld News reports that Anthropic has proposed a new AI policy framework aimed at establishing legal rules for the release of high-risk AI systems. The company hopes that governments worldwide will create regulations for cutting-edge models and prepare for AI's economic impact. Its plan covers dangerous deployment, independent testing, cybersecurity, and public resilience. Under its "AI Index Policy" initiative, Anthropic has put forward two proposals focusing on powerful models and economic policies. The company notes that the pace of AI development has outstripped existing policy-making systems, and governments need the authority to halt or restrict the deployment of dangerous models. The proposals also call for frontier developers to conduct model testing before release and to regularly publish risk reports.