Latest
  1. Johns Hopkins Bayview tells 3,182 patients an employee put their health data into a public AI tool
  2. China’s AI compute access expected to double in 2027, with US-designed chips about 65 per cent
  3. UK commits £100m to national AI Biosecurity Lab
  4. Moonshot AI releases Kimi K3.1 Pro with 1-million-token context window
  5. Residents in Newport and Cardiff oppose hyperscale datacentres near homes
  6. Study: Grok 4.7 picked Republicans in 92% of forced midterm answers, GPT-6.1 Sol picked Democrats in 97%
  7. Attackers exploiting the PHP flaws hand the break-in to AI agents, Unit 42 says
  8. Elvis Presley’s rightsholders sue Anthropic over training on 274 songs

Latest / Frontier models and safety

PolicyFrontierUSConfirmed

Anthropic publishes version 3 of its Responsible Scaling Policy

See how this story connects

Source: Anthropic: Responsible Scaling Policy Version 3.0

Anthropic released version 3.0 of its Responsible Scaling Policy, the voluntary framework it uses to manage catastrophic AI risks. The update separates Anthropic’s own plans from its recommendations for the industry, adds a Frontier Safety Roadmap, and introduces Risk Reports with external review. Anthropic said some parts of its original theory of change had worked and others had not.

Why it matters

Anthropic’s safety commitments now come with published risk reports and outside review, which regulators and other labs can compare with their own frameworks.

When
24 Feb 2026
ID
WTA-N-2026-0158
Who
Anthropic

Summary by wheresthe.ai, from Anthropic. Checked against the source on 9 Oct 2026.

Read the original at Anthropic

Source copy: download the page as saved on 18 Oct 2026 (PDF). Document the source offers: document 1 (PDF). Also held by the Internet Archive.

Cite this story

WTA-N-2026-0158: “Anthropic publishes version 3 of its Responsible Scaling Policy.” wheresthe.ai, 24 Feb 2026. https://wheresthe.ai/d/anthropic-responsible-scaling-policy-v3-2026-02/ (source: Anthropic: Responsible Scaling Policy Version 3.0).

Spotted an error? Email [email protected].