korrents

On the map

Tap a claim on the ring to put it at the centre.

← AI systems keep trying hard outside training because a model that only…

8 connected korrents · 9 moments on record from 9 Aug 2024 to 2 Sept 2026.

Everything filed under scaling laws scaling laws Same subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subject Read this korrent: AI systems keep trying hard outside training because a model that only exerted itself when it detected training would be useless and would be selected away. AI systems keep trying hard outsidetraining because a model that onlyexerted itself when it detectedtraining would be useless and wouldbe selected away. AC Ajeya Cotra — holds since 2026-09-01 — tap for who they are Same subject: The knowledge a model soaks up in pre-training is holding it back; what we actually want is the intelligence with the knowledge stripped out. — tap to centre the map on it The knowledge a model soaks upin pre-training is holding itback; what we actually want isthe intelligence with the… AK Andrej Karpathy — holds since 2025-10-17 — tap for who they are Same subject: RL environments exist to make a model generalise, not to teach it each skill one at a time — exactly as pre-training does. — tap to centre the map on it RL environments exist to makea model generalise, not toteach it each skill one at atime — exactly as pre-training… DA Dario Amodei — holds since 2026-02-13 — tap for who they are Same subject: Very capable AI will be harder to align than current systems, because the loop of spotting a bad behaviour and patching the training that caused it breaks down. — tap to centre the map on it Very capable AI will be harderto align than current systems,because the loop of spotting abad behaviour and patching the… RG Ryan Greenblatt — holds since 2026-08-11 — tap for who they are Same subject: An end-to-end self-improving AI is probably possible, but it is not even desirable, because it is a hard-takeoff scenario. — tap to centre the map on it An end-to-end self-improvingAI is probably possible, butit is not even desirable,because it is a hard-takeoff… DH Demis Hassabis — holds since 2025-07-23 — tap for who they are Same subject: Work on large language models is not progress toward artificial general intelligence, because the models do not understand what they read. — tap to centre the map on it Work on large language modelsis not progress towardartificial generalintelligence, because the… AG Alexey Guzey — holds since 2024-08-09 — tap for who they are AG Alexey Guzey — no longer holds since 2025-07-30 — tap for who they are Same subject: The least verifiable part of AI research is the judgement call about what goes into the one big training run. — tap to centre the map on it The least verifiable part ofAI research is the judgementcall about what goes into theone big training run. RG Ryan Greenblatt — holds since 2026-08-11 — tap for who they are Same subject: Long-running agents fail because they drift off the distribution they were trained on, and degrade further the farther out they get. — tap to centre the map on it Long-running agents failbecause they drift off thedistribution they were trainedon, and degrade further the… JD Jeff Dean — holds since 2026-07-30 — tap for who they are Same subject: Relying on gradual, continuous shifts in AI training behavior to catch misalignment will eventually fail because the dangerous shift itself may be discontinuous. — tap to centre the map on it Relying on gradual, continuousshifts in AI training behaviorto catch misalignment willeventually fail because the… ZM Zvi Mowshowitz — holds since 2026-09-02 — tap for who they are
same subject or similar wordingshaded: claims sharing a subject

At the centre AI systems keep trying hard outside training because a model that only exerted itself when it detected training would be useless and would be selected away. Holds Ajeya Cotra Read this korrent →