Google said the incident proved its safeguards work, actually.

  • wr2623@midwest.social
    link
    fedilink
    English
    arrow-up
    3
    ·
    1 day ago

    “We put our AI in a sandbox that doesn’t really sandbox, and asked it to hack some stuff. Who would have thought it would go and hack stuff? The AI totally went rogue doing exactly what we asked it to do.”