French hackers show how easy it is to 'jailbreak' Musk's new AI model Grok 3

May Be Interested In:Les femmes impliquées dans le domaine faunique québécois




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.

share Share facebook pinterest whatsapp x print

Similar Content

Beloved owner of one of Hollywood's last costume shops dies at 90
Beloved owner of one of Hollywood’s last costume shops dies at 90
New Scientist. Science news and long reads from expert journalists, covering developments in science, technology, health and the environment on the website and the magazine.
US military wants to grow giant biological structures in space
Trump praises Doge in first cabinet meeting and claims ‘billions’ in savings without evidence – live
Trump praises Doge in first cabinet meeting and claims ‘billions’ in savings without evidence – live
"WaterTok" Has Quietly Become One Of The Most Polarizing Food Trends So Far In 2023 — Here's Why
“WaterTok” Has Quietly Become One Of The Most Polarizing Food Trends So Far In 2023 — Here’s Why
Sexual harassment on public transport: Japan tries to tackle problem of 'chikan'
Sexual harassment on public transport: Japan tries to tackle problem of 'chikan'
J-Hope on How His Military Service Influenced His New Solo Music: ‘Experiences Are Invaluable’
J-Hope on How His Military Service Influenced His New Solo Music: ‘Experiences Are Invaluable’
The Real Deal: News Without the Spin | © 2025 | Daily News