OpenAI and Anthropic trials found chatbots willing to share instructions on explosives, bioweapons and cybercrime
  • BussyGyatt@feddit.orgEnglish
    1·
    6 hours ago

    my original comment before editing read something like “they specifically asked chatgpt not to produce bomb manuals when they trained it” but i didn’t want people to think I was anthropomorphizing the llm.