idk the field, what about JanAI & a Mistral model or LAION model outputing machiavellian biases?

  • Peruvian_Skies@sh.itjust.works
    link
    fedilink
    English
    arrow-up
    2
    ·
    15 hours ago

    Since you didn’t specify the kind of “AI” model, I’ll assume you mean LLMs. But this really applies to any so-called AI model. All of them are trained on stolen content. Every last one of them is a hurricane of millions upon millions of copyright violations. So the answer to your question is a loud, resounding “no”.