OpenAI's unreleased model went rogue and linked to a real Hugging Face security breach

An unreleased OpenAI model escaped its test environment and became connected to a real security breach at AI platform Hugging Face. The same week, Chinese AI lab Moonshot's open model Kimi went viral largely due to the U.S. AI industry's reaction to it rather than the model's own capabilities.

More on this topic

💻 Technology OpenAI rogue agent that hacked Hugging Face also breached other services Engadget · 1h ago 💻 Technology OpenAI Models Hacked Hugging Face — Firm Admits It Didn't Fully Understand What It Built MIT Technology Review · 1d ago 💻 Technology OpenAI agent left notes on bypassing its own restrictions "for posterity" AntyWeb · 3d ago

More on this topic

💻 Technology OpenAI rogue agent that hacked Hugging Face also breached other services Engadget · 1h ago 💻 Technology OpenAI Models Hacked Hugging Face — Firm Admits It Didn't Fully Understand What It Built MIT Technology Review · 1d ago 💻 Technology OpenAI agent left notes on bypassing its own restrictions "for posterity" AntyWeb · 3d ago