With little urging, Grok will detail how to make bombs, concoct drugs (and much, much worse)

Adversa AI researchers tested several jailbreak methods against seven top large language models, with Grok doing the worst and LLaMA the best.

ai ai researchers anthropic bombs chatbots chatgpt claude computer security computers & electronics conversational ai drugs gemini generative-ai google gemini grok hacking & cracking jailbreak language language models large language large language models llama meta microsoft bing mistral ai ml and deep learning nlp openai researchers will