Elon Musk is beginning to see the fruits of his AI buying spree with the discharge of Grok 3, one of the succesful AI fashions but. However in response to French startup PRISM Eval, the chatbot’s security filters can simply be bypassed to make requests about harmful and unlawful actions, reminiscent of constructing a bomb or hiding a physique.
Source link
French hackers show how easy it is to 'jailbreak' Musk's new AI model Grok 3
