Responsible Scaling Policy
AI overview
Sign in and the AI will write an overview from our coverage.
Headlines · 3
- METR says Anthropic's internal AI progress rate is nearly twice as fast
METR estimated that progress on clearly defined tasks at Anthropic reached about 1.9 times its previous rate, with a 95% confidence interval of 1.5 to 2.8. METR said this means roughly two years of capability progress in one year—not doubled model capability or uniformly doubled employee productivity—and that AI-assisted research remains primarily human-assisted. Anthropic separately disclosed that it restricted certain capabilities in Claude Fable 5.2 and Mythos 5.2 that could directly accelerate AI development, while warning that its restrictions’ effectiveness has not been independently established.
DoNews · 🔥 0 - Anthropic Says Its RSI Threshold May Be Imminent
Anthropic’s amended S-1/A says its internal estimates for future models indicate it may soon meet its Responsible Scaling Policy threshold for “dramatically accelerated progress attributable to AI,” a policy area associated with recursive self-improvement. TechCrunch reported that the filing also says AI-assisted development has previously produced capability gains the company did not recognize in time, while no current model has been identified as crossing the threshold and the supporting data remains under evaluation.
TechCrunch AI · 🔥 0 - Amodei defends Anthropic safety strategy amid AI bill dispute
Anthropic CEO Dario Amodei defended the company’s “race to the top” safety strategy in a long post responding to criticism from Encode’s Nathan Calvin. He acknowledged three communication failures, including Anthropic’s handling of a 37-page federal AI bill, and said the company withheld Mythos-class models from general availability for about four months while stronger cyber safeguards were prepared. The Next Web reported the response, while 虎嗅 AI cited Politico and Reuters reporting that the draft set a threshold of more than 50 deaths or injuries, or $1 billion in losses, for certain government model-blocking actions.
The Next Web · 🔥 0
Experience and discussion from the community
Share my Responsible Scaling Policy experienceAsk about Responsible Scaling Policy
Nobody has shared their experience with Responsible Scaling Policy yet.