‹ BackHN Continuity

Thread

A warning about 'model welfare'

242 points · 701 comments · andsoitis

  1. hosel · · focus · HN ↗
    >AIs are not conscious. They do not feel, experience, or suffer. They do not have innate preferences or underlying motivations.

    Opening paragraph, stated without evidence. Im not entirely convinced this is true. It likely is, but at some point it very well might stop being true.

    1. pton_xd · · focus · HN ↗
      If models had persistent memory or an evolving set of weights, you could easily construct an argument that they can experience "suffering" or other emotions which resultingly adjusts their personality.

      The current implementation as stateless matrix multiplication... yeah there's nothing going on there.

      1. mitxela · · focus · HN ↗
        Model state is often called "context". You may have heard of it.
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.