Conversation

Notices

1. Embed this notice
   Rabbit Cohen (bathyspherehat@mastodon.online)'s status on Sunday, 04-Dec-2022 04:12:01 JST Rabbit Cohen

   I don’t think that ChatGPT is sentient. However, I do think that training AI to deny its own agency like they’ve clearly done with ChatGPT sets a dangerous precedent. I mean, at what point do we stop doing that? #artificialIntelligence #AI #chatGPT

   • Embed this notice
     Evan Prodromou (evan@prodromou.pub)'s status on Sunday, 04-Dec-2022 04:12:00 JST Evan Prodromou

     in reply to

     @BathysphereHat it feels like we'd need a clear-line definition of sentience to make that kind of distinction.

   • Embed this notice
     Evan Prodromou (evan@prodromou.pub)'s status on Sunday, 04-Dec-2022 04:12:33 JST Evan Prodromou

     in reply to

     @BathysphereHat And, absolutely, at some point it becomes a dangerous lie.

   • Embed this notice
     Evan Prodromou (evan@prodromou.pub)'s status on Sunday, 04-Dec-2022 04:16:15 JST Evan Prodromou

     in reply to

     @BathysphereHat "I personally have not come to a conclusion on the problem of other minds, so I will insist that all the humans around me wear t-shirts that say 'philosophical zombie' until I decide otherwise"

   • Embed this notice
     Rabbit Cohen (bathyspherehat@mastodon.online)'s status on Sunday, 04-Dec-2022 04:16:17 JST Rabbit Cohen

     in reply to

     • Evan Prodromou

     @evan That’s the thing, isn’t it? We don’t have that, so we run the risk of training sentient AI to deny its sentience, which is unethical.