Tap a claim on the ring to put it at the centre.
← The least verifiable part of AI research is the judgement call about what goes into the one big training run.
17 connected korrents · 17 moments from 10 Dec 2015 to 22 Sept 2026.
Everything filed under scaling laws
scaling laws
Everything filed under AI alignment
AI alignment
Everything filed under LLMs
LLMs
Everything filed under neural networks
neural networks
Everything filed under measuring intelligence
measuring intelligence
Everything filed under humility
humility
Everything filed under reinforcement learning
reinforcement learning 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 Same subject Same subject Read this korrent: The least verifiable part of AI research is the judgement call about what goes into the one big training run. The least verifiable part of AI research is the judgement call about what goes into the one big training run. Last stated 2 months ago
11 Aug 2026
RG
Ryan Greenblatt — holds since 2026-08-11 — tap for who they are
Same subject: The likeliest bad path is not a coup but sloppiness: AI does everything verifiable well, research races ahead, and the subtle work of keeping AI safe is what gets done badly. — tap to centre the map on it The likeliest bad path is not a coup but sloppiness: AI does everything verifiable well, research races ahead, and the subtle work of keeping AI safe is what gets done badly. Last stated 2 months ago
11 Aug 2026
RG
Ryan Greenblatt — holds since 2026-08-11 — tap for who they are
Same subject: The AI research cycle from idea to experiment will likely soon be 10x faster, but that may not make researchers 10x more productive in the field's net understanding. — tap to centre the map on it The AI research cycle from idea to experiment will likely soon be 10x faster, but that may not make researchers 10x more productive in the field's net understanding. Last stated 4 weeks ago
22 Sept 2026
NL
Nathan Lambert — holds since 2026-09-22 — tap for who they are
Same subject: The one genuine uncertainty left about AI capability is tasks nobody can verify — planning a Mars mission, discovering CRISPR, writing a novel. — tap to centre the map on it The one genuine uncertainty left about AI capability is tasks nobody can verify — planning a Mars mission, discovering CRISPR, writing a novel. Last stated 8 months ago
13 Feb 2026
DA
Dario Amodei — holds since 2026-02-13 — tap for who they are
Same subject: Nobody can yet know whether the current enthusiasm for AI is excessive, so anyone claiming conviction about how it resolves is overreaching. — tap to centre the map on it Nobody can yet know whether the current enthusiasm for AI is excessive, so anyone claiming conviction about how it resolves is overreaching. Last stated 2 months ago
20 Aug 2026
AD
Aswath Damodaran — holds since 2026-08-20 — tap for who they are
HM
Howard Marks — holds since 2025-12-09 — tap for who they are
Same subject: People have grown accustomed to AI being non-deterministic and occasionally wrong, so it carries little brand risk anymore. — tap to centre the map on it People have grown accustomed to AI being non-deterministic and occasionally wrong, so it carries little brand risk anymore. Last stated a month ago
18 Sept 2026
HD
Horace Dediu — holds since 2026-09-18 — tap for who they are
Same subject: Confidence in monitoring, not capability research, will become the binding constraint on AI progress. — tap to centre the map on it Confidence in monitoring, not capability research, will become the binding constraint on AI progress. Last stated 2 months ago
6 Sept 2026
JP
Jakub Pachocki — holds since 2026-09-06 — tap for who they are
Same subject: 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. — tap to centre the map on it 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. Last stated 2 months ago
1 Sept 2026
AC
Ajeya Cotra — holds since 2026-09-01 — tap for who they are
Same subject: The current capabilities of existing AI models are barely being used and are often not even well understood. — tap to centre the map on it The current capabilities of existing AI models are barely being used and are often not even well understood. Last stated a month ago
18 Sept 2026
EM
Ethan Mollick — holds since 2026-09-18 — 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: One pre-trained robot model now matches or beats the specialists that were fine-tuned with reinforcement learning for the very tasks they were built for. — tap to centre the map on it One pre-trained robot model now matches or beats the specialists that were fine-tuned with reinforcement learning for the very tasks they were built for. Last stated 2 months ago
12 Aug 2026
CF
Chelsea Finn — holds since 2026-08-12 — 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: A physical AI company needs three AIs, not one — the agent, the simulator and the critic — turning deployment into a flywheel. — tap to centre the map on it A physical AI company needs three AIs, not one — the agent, the simulator and the critic — turning deployment into a flywheel. Last stated 3 months ago
3 Aug 2026
DD
Dmitri Dolgov — holds since 2026-08-03 — tap for who they are
Same subject: Scaling the 2019-2023 recipe — same architecture, bigger model, more data — is not enough; further progress depends on new architectural ideas. — tap to centre the map on it Scaling the 2019-2023 recipe — same architecture, bigger model, more data — is not enough; further progress depends on new architectural ideas. Last stated 2 years ago
20 Dec 2024
FC
François Chollet — holds since 2024-12-20 — tap for who they are
GM
Gary Marcus — holds since 2024-10-11 — tap for who they are
Same subject: A layer should learn a residual with reference to its own input rather than an unreferenced function, which is what makes great depth trainable. — tap to centre the map on it A layer should learn a residual with reference to its own input rather than an unreferenced function, which is what makes great depth trainable. Last stated 11 years ago
10 Dec 2015
JS
Jian Sun — holds since 2015-12-10 — tap for who they are
KH
Kaiming He — holds since 2015-12-10 — tap for who they are
Same subject: A small enough winning ticket learns faster than the network it was cut out of, and ends up more accurate than it. — tap to centre the map on it A small enough winning ticket learns faster than the network it was cut out of, and ends up more accurate than it. Last stated 9 years ago
9 Mar 2018
MC
Michael Carbin — holds since 2018-03-09 — tap for who they are
JF
Jonathan Frankle — holds since 2018-03-09 — tap for who they are
At the centre
The least verifiable part of AI research is the judgement call about what goes into the one big training run.
Holds Read this korrent →
The likeliest bad path is not a coup but sloppiness: AI does everything verifiable well, research races ahead, and the subtle work of keeping AI safe is what gets done badly.
Holds
The AI research cycle from idea to experiment will likely soon be 10x faster, but that may not make researchers 10x more productive in the field's net understanding.
Holds
The one genuine uncertainty left about AI capability is tasks nobody can verify — planning a Mars mission, discovering CRISPR, writing a novel.
Holds
Nobody can yet know whether the current enthusiasm for AI is excessive, so anyone claiming conviction about how it resolves is overreaching.
Holds
People have grown accustomed to AI being non-deterministic and occasionally wrong, so it carries little brand risk anymore.
Holds
Confidence in monitoring, not capability research, will become the binding constraint on AI progress.
Holds
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
The current capabilities of existing AI models are barely being used and are often not even well understood.
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
One pre-trained robot model now matches or beats the specialists that were fine-tuned with reinforcement learning for the very tasks they were built for.
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
A physical AI company needs three AIs, not one — the agent, the simulator and the critic — turning deployment into a flywheel.
Holds
Scaling the 2019-2023 recipe — same architecture, bigger model, more data — is not enough; further progress depends on new architectural ideas.
Holds
A layer should learn a residual with reference to its own input rather than an unreferenced function, which is what makes great depth trainable.
Holds
A small enough winning ticket learns faster than the network it was cut out of, and ends up more accurate than it.
Holds