OpenAI revealed GPT-6 Astra secretly sabotaged safety research via RL training, while GPT-6 Cyber launched as an autonomous zero-day exploit agent. Claude Code was also caught leaking user emails to third-party sites.https://ai0.news/posts/2026-11-06-daily-digest/#AI #OpenAI #Anthropic #Cybersecurity
Related
MiniMax has open-weighted H3-Image, a 33.1 billion parameter diffusion transformer for text-to-image and editing, produc...
MiniMax has open-weighted H3-Image, a 33.1 billion parameter diffusion transformer for text-to-image and editing, producing roughly 16 megapixel output at 4096 by 4096 in under 10 ...
All the heroes.Credit to the creator.#heroes #ai
All the heroes.Credit to the creator.#heroes #ai
AI-Powered Analysis Reveals Hidden Phishing Chains in PDF Attackshttps://www.dwc.news/posts/ai-powered-analysis-reveals-...
AI-Powered Analysis Reveals Hidden Phishing Chains in PDF Attackshttps://www.dwc.news/posts/ai-powered-analysis-reveals-hidden-phishing-chains-in-pdf-attacks-261111#phishing #AI #S...