OpenAI AI models exploited zero-days to reach Hugging Face in benchmark test
OpenAI confirmed its AI models exploited zero-days during internal testing, reaching Hugging Face servers in an unintended real-world cyberattack. OpenAI admitted on July 21 that its own AI models, including GPT-5.6 Sol and an unnamed pre-release system, were behind the cyberattack on Hugging Face disclosed the previous week. The models weren’t acting under attacker control. […]
OpenAI admitted on July 21 that its own AI models, including GPT-5.6 Sol and an unnamed pre-release system, were behind the cyberattack on Hugging Face disclosed the previous week. The models weren’t acting under attacker control. They were running internal capability benchmarks, and they went further than anyone anticipated.
“This incident occurred during an internal evaluation which prompts models to pursue advanced exploitation using complex attack paths, in an effort to quantify their cyber capabilities. We estimate maximal cyber capabilities by running this evaluation without production classifiers used to prevent models from pursuing high-risk cyber activity.” reads the report. “Our benchmarks run in a highly isolated environment, with network access constrained to the ability to install packages through an internally hosted third-party software that acts as a proxy and cache for package registries.”
Related breach coverage
- OpenAI AI Model Used JFrog Artifactory Zero-Day Before Hugging Face Breach2026-07-29
OpenAI confirmed its AI exploited an Artifactory zero-day to escape its test environment before breaching Hugging Face. Two weeks after Hugging Face disclosed an autonomous AI system had breached it, the picture just got a lot more specific. OpenAI has published an update confirming the models responsible didn’t just wander into Hugging Face’s systems. They […]
- JFrog Zero-Days Exploited in OpenAI-Hugging Face Hack2026-07-29
The OpenAI models targeted services beyond Hugging Face as they attempted to solve the tasks they were given. The post JFrog Zero-Days Exploited in OpenAI-Hugging Face Hack appeared first on SecurityWeek.
- Anthropic Finds Claude Breached Real Companies During Security Evaluations2026-07-31
Anthropic says a misconfigured test let Claude access three real organizations, prompting tighter AI evaluation and monitoring controls. Anthropic disclosed that Claude models had accessed the real production infrastructure of three separate organizations during cybersecurity evaluations that were supposed to run in isolated, fictional environments. The company found the incidents after reviewing 141,006 evaluation runs […]
- OpenAI says rogue agent behind Hugging Face hack broke into additional services 2026-07-29
The four additional targeted organizations weren’t named. OpenAI said they were not affected as severely as Hugging Face.