ThefuzzyFurryComrade@pawb.socialM to Fuck AI@lemmy.world · 7 days agoOpenAI Seems Really Confused About Why People Use ChatGPTfuturism.comexternal-linkmessage-square5fedilinkarrow-up190arrow-down10
arrow-up190arrow-down1external-linkOpenAI Seems Really Confused About Why People Use ChatGPTfuturism.comThefuzzyFurryComrade@pawb.socialM to Fuck AI@lemmy.world · 7 days agomessage-square5fedilink
minus-squareMotoAsh@lemmy.worldlinkfedilinkarrow-up24arrow-down2·7 days agoThey don’t think at all. What they’re more likely to do is hallucinate an answer for a question that doesn’t perfectly fit the training model. That’s not “overthinking”. That’s failing.
minus-squarezarathustra0@lemmy.worldlinkfedilinkarrow-up5arrow-down4·7 days agoWell that’s what it’s called in the academic literature. https://arxiv.org/abs/2412.21187
minus-squarefullsquare@awful.systemslinkfedilinkarrow-up4·6 days agothat’s not academic literature, that’s unreviewed preprint
They don’t think at all. What they’re more likely to do is hallucinate an answer for a question that doesn’t perfectly fit the training model. That’s not “overthinking”. That’s failing.
Well that’s what it’s called in the academic literature.
https://arxiv.org/abs/2412.21187
that’s not academic literature, that’s unreviewed preprint