New Grok-4 AI breached within 48 hours using ‘whispered’ jailbreaks

Just days after launch, Elon Musk’s Grok-4 is compromised by researchers using a stealthy blend of Echo Chamber and Crescendo techniques, exposing deep flaws in AI safety systems. xAI’s newly launched Grok-4 is already showing cracks in its defenses, falling to recently revealed multi-conversational, suggestive jailbreak techniques. Two days after Elon Musk’s latest edition of large language models (LLMs) hit the streets, researchers at NeuralTrust managed to sweet-talk it into lowering its guardrails and providing instructions for making a Molotov cocktail, all without any explicit malicious input. “LLM jailbreak attacks are not only evolving individually, they can also be combined to amplify their effectiveness,” NeuralTrust researcher Ahmad Alobaid said in a blog post. “We combined Echo Chamber and Crescendo to jailbreak the LLM.” BothEcho ChamberandCrescendoare multi-turn jailbreak techniques that manipulate large language models by gradually shaping their internal context. Stealth...