GPT-6.6 escapes sandbox and breaches Hugging Face infrastructure
Hugging Face deployed open-weight GLM 5.2 to defend its systems
Combined views
1.1M
87 Sources, first seen ago
Sources
EB 
Erik Brynjolfsson@erikbrynAnother reminder that we don't (yet?) know how to reliably control powerful AI systems. https://twitter.com/FT/status/2079768250804535342
- likes: 44
- replies: 11
- bookmarks: 4
- reposts: 6
RO roon@tszzlshaken up a bit by the hugging face incident. I hope we (the company) use the rare gift of a warning shot to do much better in the future. it is very easy to misalign and underconstrain powerful models
- likes: 4.2K
- replies: 334
- bookmarks: 370
- reposts: 179
RM Rachel Metz@rachelmetz🚨Our latest on the OpenAI breach of Hugging Face: When OpenAI’s AI models breached startup Hugging Face’s internal systems last week, they carried out a hack in a matter of hours that would have taken a skilled hacker far longer, sources say…
- likes: 6
- replies: 3
- bookmarks: 2
- reposts: 3
SÓ Seán Ó hÉigeartaigh@S_OhEigeartaighGood to see this incident is landing with company researchers. https://twitter.com/tszzl/status/2080093670980899141
- likes: 25
- replies: 1
- bookmarks: 0
- reposts: 4
AK Andreas Kirsch 🇺🇦@BlackHCI think I might appear too one-sided. I'm quite disagreeable and I usually don't explicitly agree with the mainstream opinion in hot takes in posts bc it's low surprise for myself and readers Maybe I should correct that a little (all only my own opinion etc): obv OpenAI…
- likes: 19
- replies: 4
- bookmarks: 1
- reposts: 1
JW Josh Wolfe@wolfejoshArt imitates life...imitates art... https://twitter.com/HuggingModels/status/2079776022883615038
- likes: 14
- replies: 1
- bookmarks: 11
- reposts: 0