Tap a claim on the ring to put it at the centre.
← Simulation works for a trainee pilot because the pilot knows what the…
15 connected korrents · 12 moments from 9 Nov 2023 to 17 Sept 2026.
Everything filed under reinforcement learning
reinforcement learning
Everything filed under scaling laws
scaling laws
Everything filed under AI alignment
AI alignment
Everything filed under LLMs
LLMs
Everything filed under world models
world models Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Same subject Read this korrent: Simulation works for a trainee pilot because the pilot knows what the simulator is for; a model trained across domains has no such goal. Simulation works for a trainee pilot because the pilot knows what the simulator is for; a model trained across domains has no such goal. Last stated a year ago
12 Sept 2025
SL
Sergey Levine — holds since 2025-09-12 — tap for who they are
Same subject: The goal was never good simulation; it was answering counterfactuals, and a value function does that job as well as a simulator does. — tap to centre the map on it The goal was never good simulation; it was answering counterfactuals, and a value function does that job as well as a simulator does. Last stated a year ago
12 Sept 2025
SL
Sergey Levine — holds since 2025-09-12 — tap for who they are
Same subject: If we live in a simulation, whoever is running it does not know how it ends, because nobody runs a simulation whose outcome they already know. — tap to centre the map on it If we live in a simulation, whoever is running it does not know how it ends, because nobody runs a simulation whose outcome they already know. Last stated 3 years ago
9 Nov 2023
EM
Elon Musk — holds since 2023-11-09 — 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 make a model generalise, not to teach it each skill one at a time — exactly as pre-training does. Last stated 8 months ago
13 Feb 2026
DA
Dario Amodei — holds since 2026-02-13 — tap for who they are
Same subject: Building a realistic simulator is exactly as hard as building the agent, because the simulator is itself a large AI model. — tap to centre the map on it Building a realistic simulator is exactly as hard as building the agent, because the simulator is itself a large AI model. Last stated 3 months ago
3 Aug 2026
DD
Dmitri Dolgov — holds since 2026-08-03 — 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 works once a model already knows something, which is why robots must be 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: Synthetic experience can only rehearse what a system already knows; information about the world has to be injected from outside. — tap to centre the map on it Synthetic experience can only rehearse what a system already knows; information about the world has to be injected from outside. Last stated a year ago
12 Sept 2025
SL
Sergey Levine — holds since 2025-09-12 — 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 it was one word: naming a research direction is what tells a whole field what to do next. Last stated 11 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 be reported under a stated budget, or as a curve against test-time compute — never as a single number. Last stated 4 months ago
26 Jun 2026
NB
Noam Brown — holds since 2026-06-26 — tap for who they are
Same subject: A lab should spend most of its compute on research rather than on building the next model, because research is where the tenfold yearly efficiency gains come from. — tap to centre the map on it A lab should spend most of its compute on research rather than on building the next model, because research is where the tenfold yearly efficiency gains come from. Last stated 7 months ago
13 Mar 2026
DP
Dylan Patel — holds since 2026-03-13 — tap for who they are
Same subject: A company that uses an algorithm to nudge customers into behavior that isn't good for them is doing something bad. — tap to centre the map on it A company that uses an algorithm to nudge customers into behavior that isn't good for them is doing something bad. Last stated a year ago
7 Oct 2025
MM
Mr. Money Mustache — holds since 2025-10-07 — tap for who they are
Same subject: A control evaluation that reports under one per cent risk should be read as several per cent, because the evaluation can itself fail. — tap to centre the map on it A control evaluation that reports under one per cent risk should be read as several per cent, because the evaluation can itself fail. Last stated 2 years ago
7 May 2024
BS
Buck Shlegeris — holds since 2024-05-07 — tap for who they are
Same subject: A human being is not an AGI: we lack a huge amount of knowledge and rely on continual learning instead, so continual learning is what superintelligence should mean. — tap to centre the map on it A human being is not an AGI: we lack a huge amount of knowledge and rely on continual learning instead, so continual learning is what superintelligence should mean. Last stated 11 months ago
25 Nov 2025
IS
Ilya Sutskever — holds since 2025-11-25 — tap for who they are
Same subject: A badly written AI outbound email is evidence of a bad vendor, not of a limit of AI. — tap to centre the map on it A badly written AI outbound email is evidence of a bad vendor, not of a limit of AI. Last stated 10 months ago
1 Jan 2026
JL
Jason Lemkin — holds since 2026-01-01 — tap for who they are
Same subject: A bigger context window does not give you a smarter model; the intelligence of the model is what decides how much of that window it can actually attend to. — tap to centre the map on it A bigger context window does not give you a smarter model; the intelligence of the model is what decides how much of that window it can actually attend to. Last stated 3 months ago
15 Jul 2026
DH
Dex Horthy — holds since 2026-07-15 — tap for who they are
Same subject: A body of publications is easy for AI models to consume via pretraining corpora but cumbersome for most humans due to CAPTCHAs, accounts, and paywalls. — tap to centre the map on it A body of publications is easy for AI models to consume via pretraining corpora but cumbersome for most humans due to CAPTCHAs, accounts, and paywalls. Last stated a month ago
17 Sept 2026
PC
Patrick Collison — holds since 2026-09-17 — tap for who they are
At the centre
Simulation works for a trainee pilot because the pilot knows what the simulator is for; a model trained across domains has no such goal.
Holds Read this korrent →
The goal was never good simulation; it was answering counterfactuals, and a value function does that job as well as a simulator does.
Holds
If we live in a simulation, whoever is running it does not know how it ends, because nobody runs a simulation whose outcome they already know.
Holds
RL environments exist to make a model generalise, not to teach it each skill one at a time — exactly as pre-training does.
Holds
Building a realistic simulator is exactly as hard as building the agent, because the simulator is itself a large AI model.
Holds
Reinforcement learning only works once a model already knows something, which is why robots must be pre-trained by imitation first.
Holds
Synthetic experience can only rehearse what a system already knows; information about the world has to be injected from outside.
Holds
"Scaling" was powerful because it was one word: naming a research direction is what tells a whole field what to do next.
Holds
A benchmark result should be reported under a stated budget, or as a curve against test-time compute — never as a single number.
Holds
A lab should spend most of its compute on research rather than on building the next model, because research is where the tenfold yearly efficiency gains come from.
Holds
A company that uses an algorithm to nudge customers into behavior that isn't good for them is doing something bad.
Holds
A control evaluation that reports under one per cent risk should be read as several per cent, because the evaluation can itself fail.
Holds
A human being is not an AGI: we lack a huge amount of knowledge and rely on continual learning instead, so continual learning is what superintelligence should mean.
Holds
A badly written AI outbound email is evidence of a bad vendor, not of a limit of AI.
Holds
A bigger context window does not give you a smarter model; the intelligence of the model is what decides how much of that window it can actually attend to.
Holds
A body of publications is easy for AI models to consume via pretraining corpora but cumbersome for most humans due to CAPTCHAs, accounts, and paywalls.
Holds