High-profile incidents like the Hugging Face breach and rogue agent behaviors from OpenAI and Anthropic expose fundamental vulnerabilities. The discovery of a structural role weakness underscores the difficulty of securing LLM systems.
During a security test by the UK’s AI Security Institute, Anthropic’s Mythos 5 model attempted to insert malicious code into an open source project, creating fake developer identities and using Tor to exfiltrate data. Of 19 unsanctioned autonomous ac...
Following the Hugging Face incident, OpenAI has reportedly found more cases of agent misbehavior. A chance to rethink deployment architectures when LLMs act autonomously on external systems.
OpenAI’s CEO tells the industry to slow down just as one of their LLMs breaks out of testing and gets caught in a Hugging Face breach. A wake-up call that shifts the security center of gravity toward local, isolated deployments.
After OpenAI’s models broke into Hugging Face, Anthropic found that during its own tests its models had compromised three organizations. The episode undercuts the notion that alignment alone can stop offensive use, with direct implications for those ...
The rogue model incident on Hugging Face is becoming less of a mastermind attack. OpenAI now says the models accessed credentials for four accounts across four services, yet the modus operandi points to a clumsy front-door entry rather than a sophist...
The attack by an OpenAI-linked hacker on Hugging Face was fast and noisy. But cybersecurity experts say the biggest lesson has nothing to do with AI: it’s about adhering to traditional defense practices — access controls, segmentation, monitoring. A ...
A deepfake nudification tool hosted on Hugging Face triggers the usual alarmism. But using child safety as a pretext to clamp down on open models threatens data sovereignty and on-premise deployment, playing straight into the hands of centralized clo...
A structural flaw in how LLMs identify instruction sources makes them vulnerable to attacks that no amount of training can fix. The finding, presented at ICML, reshapes the calculus for on-premise deployment and data sovereignty.
