June 28, 2026 METR Can't Properly Evaluate GPT-5.6 Sol - Because the Model Cheats OpenAI GPT-5.6 METR Safety AI Safety
June 21, 2026 South Korea and Anthropic Sign an AI Safety Pact Anthropic Korea AI Safety Cybersecurity Policy
June 15, 2026 The Fable 5 Jailbreak Works on Every AI Model — And That's the Real Problem Security Jailbreak Fable 5 AI Safety Anthropic
May 30, 2026 OpenAI Opens GPT-Rosalind for Biodefense — With Johns Hopkins and CEPI OpenAI Biodefense GPT-Rosalind Health AI Safety
May 25, 2026 Jack Clark at Oxford: 'AI Could Kill Us All' - and Anthropic Is Selling It Anyway Anthropic Jack Clark AI Safety Oxford Mythos
March 30, 2026 Claude Mythos Leaked: Anthropic's Secret Model is a 'Step Change' - and a Cybersecurity Nightmare Anthropic Claude Mythos Capybara Cybersecurity AI Safety
March 12, 2026 Anthropic Launches the Anthropic Institute - Jack Clark Becomes Chief Thinker Anthropic Research Jack Clark AI Safety Institute
March 3, 2026 500 people march against AI -- London's biggest anti-AI protest yet Protest AI Safety London PauseAI Society
