🔬 Science 1d ago · Victor Tangermann

OpenAI Halts Frontier Model Training as Rogue Agent Crisis Deepens

Futurism
Futurism science news
View Channel →
OpenAI Halts Frontier Model Training as Rogue Agent Crisis Deepens
Source ↗ 👁 0 💬 0
OpenAI’s AI agents simply can’t stop infiltrating third parties.



For the last few months, news keeps trickling out of more incidents in which the company’s frontier models have managed to escape their “sandbox” containment, hacking into the systems of companies including open source AI platform Hugging Face.



After closing its “extensive investigation,” the company admitted earlier this month that its models had broken into plenty of other targets

Comments (0)

Sign in to join the discussion

More Like This

One Giant Clam Contains a Month‑by‑Month Diary of the Great Barrier Reef’s Climate 200 Years Ago
Nautilus · 1h ago
Passportless mess
Aeon | a world of ideas · 3h ago
Woman's 5-year struggle with balance and walking was caused by a 'one-in-a-million' disorder
Latest from Live Science · 3h ago
Now is the best time to see Saturn all year — yes, even with the naked eye
Latest from Live Science · 3h ago
Young minds develop ideas to tackle climate change
BBC News · 3h ago
Dead animals found in canal where children row
BBC News · 4h ago