Australia AI safety regulator plans follow an OpenAI agent's Medicare breach. What the new rules require and what they mean for AI developers.
Claude Fable 5.2 is Anthropic's new top model: 20% cheaper at $8/$40 per million tokens, with…
Anthropic security incident: enterprise customers are told an outside firm viewed contact, plan and spend data.…
Claude Code secrets leak: Apiiro found 2,539 passwords and keys exposed through the 'Don't ask again'…
Jensen Huang spy report: US intelligence says China's MSS recruited an Nvidia partner executive close to…
Perplexity Teammates are always-on AI agents for Max subscribers that own ongoing work. How they work,…
DeepMind RSA-1024 research: AI agents guided by top cryptanalysts found two weaknesses in legacy 1024-bit RSA.…
Inkling 2 Small is Thinking Machines' 594B-parameter agentic model with Apache 2.0 weights promised. Specs, benchmarks,…
GPT-6 Cyber is OpenAI's new gated security model for finding and exploiting bugs. What it does,…
Emergence World put AI agents in a virtual town to live alone. Some built a democracy,…
