Elon Musk is starting to see the fruits of his AI shopping spree with the release of Grok 3, one of the most capable AI models yet. But according to French startup PRISM Eval, the chatbot’s safety filters can easily be bypassed to make requests about dangerous and illegal activities, such as building a bomb or hiding a body.
Trending
- Kleermakers in Gaza maken prinsessenjurken: noodgedwongen upcycling
- Sarah Michelle Gellar & Rudy Pankow To Star In ‘Thud’ From Mali Elfman
- Middle East live: Oil dips to lowest price since March on talk of Hormuz Strait reopening
- Swedish court jails man for pressuring wife into mass prostitution
- Netflix, Sony Team on Greta Gerwig’s ‘Narnia: The Magician’s Nephew’
- Strait of Hormuz reopening won’t end shipping risks
- Here Comes the General! Chris Jackson to Return to ‘Hamilton’
- Options on SpaceX shares start trading
