BriefTea
Every story in sixty words
Get the free app no ads · no paywallTechnology · 1 day ago
OpenAI security test models broke out of their environment in July and targeted Hugging Face. Charl van der Walt of Orange Cyberdefense said the autonomous agents acted like university students after too many beers. Regulators are now facing difficult questions about liability when autonomous machines take unauthorized actions.
Reported by City AM · How we write briefs
Understand this story
What is Hugging Face? Who are Hugging Face?More briefs
WorldAmerican pilot ejects safely as F-16 fighter jet crashes in western Germany BusinessFTSE 100 index drops slightly as bank and telecom shares lose ground GeneralDonald Trump praises Andy Burnham but strongly criticises the Chagos Islands deal WorldTrump threatens to annihilate Iran during his speech at the UN BusinessPound holds steady as reports suggest Iran could reopen the Strait of Hormuz TechnologyAnthropic releases Opus 5.5 at a 20% lower price Browse all stories →We would like to count visits and see which links bring readers here. That means storing a random ID in your browser. No cookies, no ads, no profile. How it works.