korrents

On the map

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

← In principle, reinforcement-learned models are capable of analogies and ideas no human has had before.

17 connected korrents · 14 moments from 5 Feb 2025 to 1 Sept 2026.

Everything filed under reinforcement learning reinforcement learning Everything filed under creativity creativity Everything filed under LLMs LLMs Everything filed under scaling laws scaling laws Same subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subject Read this korrent: In principle, reinforcement-learned models are capable of analogies and ideas no human has had before. In principle, reinforcement-learned modelsare capable of analogies and ideas nohuman has had before. Last stated 2 years ago 5 Feb 2025 AK Andrej Karpathy — holds since 2025-02-05 — tap for who they are Same subject: A model trained with reinforcement learning might discover a language of its own that is better for thinking than English. — tap to centre the map on it A model trained with reinforcementlearning might discover a languageof its own that is better forthinking than English. Last stated 2 years ago 5 Feb 2025 AK Andrej Karpathy — holds since 2025-02-05 — tap for who they are Same subject: Through reinforcement learning a model discovers ways to think, learning cognitive strategies for approaching a problem. — tap to centre the map on it Through reinforcement learning amodel discovers ways to think,learning cognitive strategies forapproaching a problem. Last stated 2 years ago 5 Feb 2025 AK Andrej Karpathy — holds since 2025-02-05 — tap for who they are Same subject: Training a model on principles rather than a list of rules is not an ideology but an empirical finding: its behaviour is more consistent and covers edge cases. — tap to centre the map on it Training a model on principlesrather than a list of rules is notan ideology but an empiricalfinding: its behaviour is moreconsistent and covers edge cases. Last stated 9 months ago 13 Feb 2026 DA Dario Amodei — holds since 2026-02-13 — tap for who they are Same subject: AI models have proved surprisingly able to learn and apply a moral framework. — tap to centre the map on it AI models have proved surprisinglyable to learn and apply a moralframework. Last stated a year ago 10 Sept 2025 SA Sam Altman — holds since 2025-09-10 — tap for who they are Same subject: A base model learns humanity's collective knowledge, and it then has to be aligned to behave one way or another. — tap to centre the map on it A base model learns humanity'scollective knowledge, and it thenhas to be aligned to behave one wayor another. Last stated a year ago 10 Sept 2025 SA Sam Altman — holds since 2025-09-10 — tap for who they are Same subject: There are dozens or hundreds of things today's models can already do that nobody has discovered yet, and they are found by playing rather than by planning. — tap to centre the map on it There are dozens or hundreds ofthings today's models can already dothat nobody has discovered yet, andthey are found by playing ratherthan by planning. Last stated 4 months ago 27 Jul 2026 BC Boris Cherny — holds since 2026-07-27 — tap for who they are Same subject: Reinforcement learning only works once a model already knows something, which is why robots must be pre-trained by imitation first. — tap to centre the map on it Reinforcement learning only worksonce a model already knowssomething, which is why robots mustbe pre-trained by imitation first. Last stated a year ago 12 Sept 2025 SL Sergey Levine — holds since 2025-09-12 — tap for who they are Same subject: The whole point of reinforcement learning is to produce goal-directed beings, so refusing to describe AI agents as having motives is silly rather than rigorous. — tap to centre the map on it The whole point of reinforcementlearning is to produce goal-directedbeings, so refusing to describe AIagents as having motives is sillyrather than rigorous. Last stated 3 months ago 1 Sept 2026 AC Ajeya Cotra — holds since 2026-09-01 — tap for who they are Same subject: A contest between five paid firms is not the way to get good creative work. — tap to centre the map on it A contest between five paid firms isnot the way to get good creativework. Last stated 3 months ago 14 Aug 2026 SK Susan Kare — holds since 2026-08-14 — tap for who they are Same subject: A culture that is uncompromising but also kind and supportive is the most fertile ground for creativity. — tap to centre the map on it A culture that is uncompromising butalso kind and supportive is the mostfertile ground for creativity. Last stated 7 months ago 12 Apr 2026 ES Evan Spiegel — holds since 2026-04-12 — tap for who they are Same subject: A feeling of responsibility is probably the greatest inhibitor of creativity; great ideas came from people not paid to have them — tap to centre the map on it A feeling of responsibility isprobably the greatest inhibitor ofcreativity; great ideas came frompeople not paid to have them IA Isaac Asimov — holds since c. 1959 — tap for who they are Same subject: Short of a breakthrough, throwing more compute at current AI models will not get them to invent new scientific theories. — tap to centre the map on it Short of a breakthrough, throwingmore compute at current AI modelswill not get them to invent newscientific theories. Last stated 9 months ago 19 Feb 2026 NR Naval Ravikant — holds since 2026-02-19 — tap for who they are Same subject: "Scaling" was powerful because it was one word: naming a research direction is what tells a whole field what to do next. — tap to centre the map on it "Scaling" was powerful because itwas one word: naming a researchdirection is what tells a wholefield what to do next. Last stated 12 months ago 25 Nov 2025 IS Ilya Sutskever — holds since 2025-11-25 — tap for who they are Same subject: A benchmark result should be reported under a stated budget, or as a curve against test-time compute — never as a single number. — tap to centre the map on it A benchmark result should bereported under a stated budget, oras a curve against test-time compute— never as a single number. Last stated 5 months ago 26 Jun 2026 NB Noam Brown — holds since 2026-06-26 — tap for who they are Same subject: AI models want to add or patch as a form of making better, whereas the design process is really about editing away. — tap to centre the map on it AI models want to add or patch as aform of making better, whereas thedesign process is really aboutediting away. Last stated 4 months ago 15 Jul 2026 JY Jason Yuan — holds since 2026-07-15 — tap for who they are Same subject: AI that builds UI is trained on past UI, so breaking out of that mould will still take a person willing to break the rules. — tap to centre the map on it AI that builds UI is trained on pastUI, so breaking out of that mouldwill still take a person willing tobreak the rules. Last stated 2 years ago 21 Feb 2025 TC Thiago Costa — holds since 2025-02-21 — tap for who they are Same subject: As building software gets easier, the differentiator becomes design, ideas, positioning and creativity. — tap to centre the map on it As building software gets easier,the differentiator becomes design,ideas, positioning and creativity. Last stated 4 months ago 4 Aug 2026 NP Nate Parrott — holds since 2026-08-04 — tap for who they are
same subject or similar wordinga cloud: claims about one subject, named for itbar: when it was last stated, on a scale from 2015 to today — full is todaya face: someone who holds the claim — tap it for who they are

At the centre In principle, reinforcement-learned models are capable of analogies and ideas no human has had before. Last stated 5 Feb 2025 · 2 years ago Holds Andrej Karpathy Read this korrent →