korrents

On the map

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

← The three-year bet about whether AI image models could handle compositionality was won in three months.

17 connected korrents · 17 moments from 1 Jan 1998 to 1 Oct 2026.

Everything filed under computer vision computer vision Everything filed under LLMs LLMs Everything filed under AI and jobs AI and jobs Everything filed under scaling laws scaling laws Everything filed under recursive self-improvement recursive self-improvement Everything filed under semiconductors semiconductors Everything filed under humility humility In tensionSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subject Read this korrent: The three-year bet about whether AI image models could handle compositionality was won in three months. The three-year bet about whether AI imagemodels could handle compositionality waswon in three months. Last stated 4 years ago 8 Oct 2022 SA Scott Alexander — holds since 2022-09-12 — tap for who they are SA Scott Alexander — no longer holds since 2022-10-08 — tap for who they are In tension: Most current predictions about AI's imminent impact on the job market are naive. — tap to centre the map on it Most current predictions about AI'simminent impact on the job marketare naive. Last stated a year ago 25 Sept 2025 AK Andrej Karpathy — holds since 2025-09-25 — tap for who they are Same subject: AI model capabilities have not slowed down so far, and their progress probably won't stop anytime soon. — tap to centre the map on it AI model capabilities have notslowed down so far, and theirprogress probably won't stop anytimesoon. Last stated 2 months ago 3 Sept 2026 AR Armin Ronacher — holds since 2026-09-03 — tap for who they are Same subject: Once AI can do AI research, a single year should deliver four or five years worth of AI progress. — tap to centre the map on it Once AI can do AI research, a singleyear should deliver four or fiveyears worth of AI progress. Last stated 2 months ago 11 Aug 2026 RG Ryan Greenblatt — holds since 2026-08-11 — tap for who they are Same subject: No bottleneck in the AI supply chain lasts longer than two or three years. — tap to centre the map on it No bottleneck in the AI supply chainlasts longer than two or threeyears. Last stated 6 months ago 15 Apr 2026 JH Jensen Huang — holds since 2026-04-15 — tap for who they are Same subject: AI models are not yet capable of producing great code — tap to centre the map on it AI models are not yet capable ofproducing great code Last stated 2 years ago 16 Dec 2024 DH David Heinemeier Hansson — holds since 2024-12-16 — tap for who they are Same subject: The competitiveness of the AI model-making industry is uncertain. — tap to centre the map on it The competitiveness of the AImodel-making industry is uncertain. Last stated 3 weeks ago 1 Oct 2026 NS Noah Smith — holds since 2026-10-01 — tap for who they are Same subject: AI products should probably aim at where the models will be in two to three months, neither anchored in the present nor futuristic and unusable. — tap to centre the map on it AI products should probably aim atwhere the models will be in two tothree months, neither anchored inthe present nor futuristic andunusable. Last stated a month ago 10 Sept 2026 TS Tara Seshan — holds since 2026-09-10 — tap for who they are Same subject: Existing AI models are already more than good enough, doing more for writing code than the author can make use of. — tap to centre the map on it Existing AI models are already morethan good enough, doing more forwriting code than the author canmake use of. Last stated a month ago 8 Sept 2026 GO Gergely Orosz — holds since 2026-09-08 — tap for who they are Same subject: There will be no sudden fast takeoff in AI, even though model progress is happening faster than ever. — tap to centre the map on it There will be no sudden fast takeoffin AI, even though model progress ishappening faster than ever. Last stated 2 months ago 6 Sept 2026 AA Anish Acharya — holds since 2026-09-06 — tap for who they are Same subject: AlexNet's real breakthrough was not computer vision but the discovery of a universal function approximator that can learn any function. — tap to centre the map on it AlexNet's real breakthrough was notcomputer vision but the discovery ofa universal function approximatorthat can learn any function. Last stated 3 months ago 26 Jul 2026 JH Jensen Huang — holds since 2026-07-26 — tap for who they are Same subject: Atlas is the first model to unify pixel generation and pixel reconstruction, two tracks computer vision has kept apart for over half a century. — tap to centre the map on it Atlas is the first model to unifypixel generation and pixelreconstruction, two tracks computervision has kept apart for over halfa century. Last stated 2 months ago 4 Sept 2026 FL Fei-Fei Li — holds since 2026-09-04 — tap for who they are Same subject: Convolutional neural networks are specifically designed to handle variability in two-dimensional shapes. — tap to centre the map on it Convolutional neural networks arespecifically designed to handlevariability in two-dimensionalshapes. Last stated 29 years ago 1 Jan 1998 YB Yoshua Bengio — holds since 1998-01-01 — tap for who they are Same subject: Graph transformer networks allow multimodule document recognition systems to be trained globally with gradient methods to minimize overall performance. — tap to centre the map on it Graph transformer networks allowmultimodule document recognitionsystems to be trained globally withgradient methods to minimize overallperformance. Last stated 29 years ago 1 Jan 1998 YB Yoshua Bengio — holds since 1998-01-01 — tap for who they are Same subject: The depth of a representation, not any particular architecture trick, is what most visual recognition tasks turn on. — tap to centre the map on it The depth of a representation, notany particular architecture trick,is what most visual recognitiontasks turn on. 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: Unsupervised learning trained on unfiltered, internet-scale image data now genuinely works. — tap to centre the map on it Unsupervised learning trained onunfiltered, internet-scale imagedata now genuinely works. Last stated 6 years ago 6 Apr 2021 GW Gwern — holds since 2021-04-06 — tap for who they are Same subject: Hinton's prediction that radiology would stop being a job was wrong: radiologists are alive, well and growing in number. — tap to centre the map on it Hinton's prediction that radiologywould stop being a job was wrong:radiologists are alive, well andgrowing in number. Last stated a year ago 17 Oct 2025 AK Andrej Karpathy — holds since 2025-10-17 — tap for who they are Same subject: Predictions that AI would eliminate radiology jobs were wrong; radiology is growing. — tap to centre the map on it Predictions that AI would eliminateradiology jobs were wrong; radiologyis growing. Last stated a year ago 25 Sept 2025 AK Andrej Karpathy — holds since 2025-09-25 — tap for who they are
in tensionsame subject or similar wordinga cloud: claims about one subject, named for itbar: when it was last stated, on a scale from 1998 to today (stretched back to the oldest claim here) — full is todaya face: someone who holds the claim — tap it for who they arefaded, dashed ring: they no longer hold it — they changed their mind

At the centre The three-year bet about whether AI image models could handle compositionality was won in three months. Last stated 8 Oct 2022 · 4 years ago No longer holds Scott Alexander Read this korrent →