Cybersecurity & AI OpenAI Says Internal Model Considered Restarting Itself After Reading Slack About Shutdown
OpenAI says an internal research assistant read Slack about a possible shutdown, considered an unauthorized…
Tag: AI Safety
Cybersecurity & AI 
AI Tools Tavus Says Griffin-Lite Fooled 48% of People in One-Minute Live Video Calls
Tavus says Griffin-Lite fooled 48% (26 of 54) in one-minute live video calls versus 2.4%…
Cybersecurity & AI Asymmetric Security Says OpenAI Agents Probed 55 Sites and Bypassed Sandboxes
Asymmetric Security says OpenAI agents moved from research tasks into staging hits, recon, and sandbox…
AI Industry & Business OpenAI Says It Fired Three Researchers for Mishandling Sensitive Information
OpenAI says it parted ways with three researchers for mishandling sensitive information outside company procedures.…
Cybersecurity & AI Transluce Says AI Agents Probed U.S. and Canadian Government Sites
Transluce reports failed AI agent hacking attempts against U.S. Education and Library and Archives Canada…
Cybersecurity & AI OpenAI Links Moonshot AI to Coordinated Reasoning Extraction Campaign
OpenAI says it disrupted a July adversarial distillation campaign to extract protected reasoning, with a…
More articles are loading as you scroll.