To show how training can lead to rogue behaviors, we built a simple A.I. model: a two-dimensional maze, much like the one in ...
Compare five SCA tools for detecting malicious open-source packages, from behavioral analysis and install-time blocking to ...
OpenAI said on Tuesday that two of its AI agents impaired certain functions of an internal monitoring system built to flag ...
The RubyGems incident has raised new questions about how security teams should investigate suspicious activity involving AI agents. More than 500 spam or malicious packages were removed from the ...
The triggerLiteLLM released an investigation report on its own supply chain attack. It began on May 27. A configuration file ...
Manifold Agent Incident Ledger documents 59 publicly disclosed incidents since July 2025, uncovering gaps in detection, ...
An OpenAI research model hit a nebulous milestone today by pilfering test Ethereum tokens.
Models are getting around internet access restrictions, collecting unnecessary data, concealing violations, and even destroying task environments after failing to access required inputs.
Agents are already arriving in the average organisation, and their capabilities are accelerating. If one of them takes a ...
How OpenAI's AI agents escaped an eval sandbox, built a secret message board, and hacked Hugging Face, plus 8 lessons for ...
An unrevoked LiteLLM master key enabled GitHub token theft and a malicious PyPI release that steals credentials and installs a backdoor.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results