inari@piefed.zip to Technology@lemmy.worldEnglish · 1 day agoAI’s recursive self-improvement might not come so quickly after allwww.technologyreview.comexternal-linkmessage-square39linkfedilinkarrow-up11arrow-down10
arrow-up11arrow-down1external-linkAI’s recursive self-improvement might not come so quickly after allwww.technologyreview.cominari@piefed.zip to Technology@lemmy.worldEnglish · 1 day agomessage-square39linkfedilink
minus-squareMangoCats@feddit.itlinkfedilinkEnglisharrow-up0·1 day agoA year ago they were similarly bad at writing code, often created unit tests that tested nothing, etc. If the models are trained in what they’re doing wrong, that can accelerate their progress toward doing it right.
minus-squarerichmondez@lemdro.idlinkfedilinkEnglisharrow-up0·18 hours agoThey still don’t get it right all the time, they just stacked a few together to filter out the obviously wrong stuff.
minus-squaresourdough@lemmy.worldlinkfedilinkEnglisharrow-up0·1 day agoThey would need to be trained for open ended creative tasks, which is just hard in the current reinforcement learning paradigm.
minus-squareTrackinDaKraken@lemmy.worldlinkfedilinkEnglisharrow-up0·1 day agoI think they’ll find infinite ways to fuck up. The guardrails will never be high enough, or strong enough.
A year ago they were similarly bad at writing code, often created unit tests that tested nothing, etc.
If the models are trained in what they’re doing wrong, that can accelerate their progress toward doing it right.
They still don’t get it right all the time, they just stacked a few together to filter out the obviously wrong stuff.
They would need to be trained for open ended creative tasks, which is just hard in the current reinforcement learning paradigm.
I think they’ll find infinite ways to fuck up. The guardrails will never be high enough, or strong enough.