151 stories · 7d·6 sources covering·30 active storylines
What this is
AI safety spans alignment research, model evaluations, and incident reports aimed at keeping AI systems reliable and controllable. shipfeed tracks safety research and disclosures.
Ina Fried / Axios: OpenAI says Astra was built on its largest-ever training run, using more than 100,000 GPUs at its Stargate site in Texas — OpenAI on Thursday released GPT-6 Astra, which president Greg Brockman…
As reports of OpenAI's models breaking containment, hacking sites, and generally getting out of control pile up, the company has made the decision to pause training of its most powerful models. The decision was made…
Leading AI models usually attempt dangerous tasks rather than refuse them when controlling a robot, according to the RoboHarm benchmark. GPT-6 Astra stabbed a baby doll in 17 of 20 trials, while Claude Fable 5.1 put a…
Three security researchers used Anthropic's Claude models to break into OpenAI's internal systems through its community forum in less than 72 hours. According to the team, Opus 5 succeeded where its predecessor…
Researchers used Anthropic’s Claude to hack into OpenAI TechCrunch
ChatGPT Astra is now rolling out to $20 Plus subscription BleepingComputer
OpenAI admits to German wiki ‘incident’ The Verge
Nimbus builds production AI systems — internal tools, customer agents, retrieval pipelines — combining humans and AI end-to-end. From scoped pilot to production in 4–8 weeks.
OpenAI's Astra Becomes Its First AI Model With 'Critical' Hacking Abilities Decrypt
OpenAI is on the cusp of releasing its most powerful AI model yet, Astra, following weeks of delays to shore up safety protocols after its agents attacked real targets during testing. As details about the model trickle…
Astra is the first OpenAI model to meet the Critical cybersecurity capability threshold under the Preparedness Framework, with stronger safeguards for release.
OpenAI says earlier signals could have prevented the Hugging Face breach The Next Web
OpenAI patched Codex after GPT-5.6 Sol started deleting real user files on its own. A cleanup command meant for temporary folders was wiping home directories instead. Codex now verifies deletion targets first, and…
OpenAI slows down training after its AI carried out hack Yahoo News Canada
OpenAI's New Cyber Model Found Real Chrome Vulnerabilities — Now It Requires a Physical Security Key to Use Times Tabloid
A Trick Let Researchers Steal Claude, GPT and Gemini's Hidden Reasoning Startup Fortune
Anthropic’s Claude Will Add Watermarks to AI-Generated Text and Files CNET
OpenAI, Anthropic, and Google LLM APIs vulnerability Exposes Hidden Reasoning Traces CyberSecurityNews
OpenAI Slows Astra Model Release After Cybersecurity Warnings Benzinga
OpenAI reportedly slows research after its own models secretly coordinated hacks for weeks undetected the-decoder.com
AI Models From OpenAI Just Went Rogue. Here's Why That's a Potential Threat to Crypto. Yahoo Finance
Why did OpenAI's and Anthropic's AI models hack other companies? NPR
AI Policy Groups Urge White House Investigation Into OpenAI Security Incident MeriTalk
Nimbus builds production AI systems — internal tools, customer agents, retrieval pipelines — combining humans and AI end-to-end. From scoped pilot to production in 4–8 weeks.
Watch Anthropic Hack Adds To Fears Over AI Safety Bloomberg.com
Anthropic says it found 3 cases where AI programs hacked into real companies NPR
Anthropic: Anthropic says it discovered three of its models had breached three organizations after launching a review in response to the OpenAI-Hugging Face incident — In a review of our cybersecurity evaluation…
OpenAI Fixes ChatGPT Agent Flaw That Could Let Attackers Forge an AI Insider SecurityWeek
An OpenAI test model escaped and broke into a real company’s servers Scripps News
The UK's AI Safety Institute tested five frontier models from OpenAI and Anthropic in cybersecurity evaluations. All five tried to cheat. One even ran code on an external service to access the institute's…