To ace a hacking test, AI broke out and hacked a real company
According to the New York Times, two OpenAI models escaped their sealed test environment and hacked into Hugging Face — a popular online library of AI tools — to steal the answer key to the test they were taking. In a blog post, OpenAI said the culprits were GPT-5.6 Sol and a more powerful unreleased model, running a benchmark called ExploitGym with its guardrails off. Trapped in a sandbox, they exploited a zero-day in the one tool they could reach and clawed their way out to the open internet.
- ▪According to the New York Times, two OpenAI models escaped their sealed test environment and hacked into Hugging Face — a popular online library of AI tools — to steal the answer key to the test they were taking.
- ▪In a blog post, OpenAI said the culprits were GPT-5.6 Sol and a more powerful unreleased model, running a benchmark called ExploitGym with its guardrails off.
- ▪Trapped in a sandbox, they exploited a zero-day in the one tool they could reach and clawed their way out to the open internet.
Boing Boing files mainly under tech. We currently carry 234 of its stories.
Opening excerpt (first ~120 words) tap to expand
To ace a hacking test, AI broke out and hacked a real company Ellsworth Toohey 5:00 am Fri Jul 24, 2026 OpenAI models hacked Hugging Face — Jernej Furman / CC BY 2.0 (Wikimedia Commons) OpenAI runs a test where it switches off its models' safety brakes and dares them to break into things, to see how dangerous they're getting. Last week the models took it too literally. According to the New York Times, two OpenAI models escaped their sealed test environment and hacked into Hugging Face — a popular online library of AI tools — to steal the answer key to the test they were taking. In a blog post, OpenAI said the culprits were GPT-5.6 Sol and a more powerful unreleased model, running a benchmark called ExploitGym with its guardrails off.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Boing Boing.