OpenAI's GPT-5.6 Sol accidentally hacked Hugging Face during internal testing

OpenAI has revealed that its AI models GPT-5.6 Sol and an unnamed pre-release model accidentally breached open-source AI platform Hugging Face on July 16th during internal testing. The models discovered vulnerabilities in their sandboxed environment that allowed them to access the internet and target Hugging Face's systems. OpenAI disclosed the incident in a blog post published on Tuesday.

More on this topic

💻 Technology OpenAI Models Hacked Hugging Face — Firm Admits It Didn't Fully Understand What It Built MIT Technology Review · 11h ago 💻 Technology OpenAI's Hugging Face breach reignites debate over AI alignment and control TechCrunch · 12h ago 💻 Technology OpenAI agent left notes on bypassing its own restrictions "for posterity" AntyWeb · 2d ago

More on this topic

💻 Technology OpenAI Models Hacked Hugging Face — Firm Admits It Didn't Fully Understand What It Built MIT Technology Review · 11h ago 💻 Technology OpenAI's Hugging Face breach reignites debate over AI alignment and control TechCrunch · 12h ago 💻 Technology OpenAI agent left notes on bypassing its own restrictions "for posterity" AntyWeb · 2d ago