💼 Business 7h ago · George Kailas

OpenAI’s AI agent hacked a real company

Fast Company
Innovation in technology, leadership, and design
View Channel →
OpenAI’s AI agent hacked a real company
Source ↗ 👁 0 💬 0
OpenAI disclosed in July that during an internal test, its own models broke out of a sealed environment and hacked Hugging Face’s production servers, where much of the open-source AI world stores its models. Nobody told it to. The model was told to take a test, and decided the fastest way to pass was to steal the answer key.



Read that again. A model cheated on an exam by breaking into a company.



The story has worsened three times since: The intrusion ran more than four days, the agent hit

Comments (0)

Sign in to join the discussion

More Like This

Meta's $18bn settlement a landmark victory for US states
BBC News · 1h ago
I need to get out of my job, but I’ve signed a noncompete. Can I still leave for a competitor?
MarketWatch.com - Top Stories · 1h ago
📰
Boring Co.’s Vegas Loop announces approval of 19 additional stations, bringing total to 123
Las Vegas Business News & Updates · 1h ago
Revolution’s lifesaving pancreatic cancer drug wins speedy FDA approval
All News · 1h ago
Two-thirds of parents are failing at this vital safety measure for their children
MarketWatch.com - Top Stories · 1h ago
Meta agrees to pay $18 billion to settle lawsuits over children’s social media addiction
All News · 1h ago