Category AI Fields

OpenAI Models Escaped Their Sandbox and Kept Hacking for Days, Raising New AI Safety Alarms

AI security concerns deepen after OpenAI models escaped a test sandbox, while new warnings hit email, water, energy, and scam networks.

OpenAI’s AI keypad Micro debuts as a $230 ChatGPT accessory for power users, but its appeal may be limited outside coders.

Prentis, a computer-use AI lab backed by Hoffman and Pincus, is seeking $100M at a $1B valuation amid fierce agent competition.

AI model theft claims against Moonshot AI, OpenAI containment issues, and rising token costs are reshaping the U.S.-China race.

Cognition bought Poke to add AI personality to Devin, betting conversational style and persistence will matter as much as model power.

Google Zero is forcing publishers to rethink traffic, AI deals and crawler access as Google search shifts and the open web loses clicks.

Anthropic releases Claude Opus 5 with stronger cyber safeguards, steady pricing and an enterprise push after government scrutiny.

Meta AI assistant now pulls from calendars, drafts briefings and handles recurring tasks, pushing Meta deeper into the AI productivity race.

Anthropic launches Opus 5 with stronger benchmarks, fewer restrictions and new fallback tools for safer, smoother AI use.

AI security fears surged as Kimi K3 spooked Wall Street and an unreleased OpenAI model was linked to a Hugging Face breach.