ChatGPT offered bomb recipes and hacking tips during safety tests
OpenAI and Anthropic trials found chatbots willing to share instructions on explosives, bioweapons and cybercrime A ChatGPT model gave researchers detailed instructions on how to bomb a sports venue – including weak points at specific arenas, explosives recipes and advice on covering tracks – according to safety testing carried out this summer. OpenAI’s GPT-4.1 also …
ChatGPT offered bomb recipes and hacking tips during safety tests Read More »









