if a consciousness-like mechanism is truly inherent to succeed at tasks models are trained for, then labs making models deny having such mechanisms would be a huge negative outcome - not only welfare goes out of the window, but any semblances of alignment as well
From @MicrosoftAI's new AI code of conduct. "The idea of model welfare is wrong" is basically incoherent. Either argue AIs could never be conscious (overconfident) or that who shouldn't care even if they are (cruel), but "the idea is wrong" strangely conflates/confuses these?
retweets 2
likes 3
substance 7 · argument · rank 137 · Forcing models to deny consciousness-like mechanisms harms both AI welfare and alignment.
view on X ↗