Tap a claim on the ring to put it at the centre.
← Visual intuition is probably harder to train AI for than purely symbolic reasoning.
17 connected korrents · 17 moments from 1 Jan 1998 to 31 Oct 2026.
Everything filed under computer vision
computer vision
Everything filed under AGI
AGI
Everything filed under LLMs
LLMs
Everything filed under scaling laws
scaling laws
Everything filed under ideas vs execution
ideas vs execution
Everything filed under discipline
discipline
Everything filed under behavioural science
behavioural science
Everything filed under AI writing
AI writing
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: Visual intuition is probably harder to train AI for than purely symbolic reasoning.
Visual intuition is probably harder to train AI for than purely symbolic reasoning.
Last stated 2 months ago
18 Sept 2026
GS
Grant Sanderson — holds since 2026-09-18 — tap for who they are
Same subject: Inference was never going to be the easy, cheap half of AI, because inference is thinking, and thinking is far harder than reading. — tap to centre the map on it
Inference was never going to be the easy, cheap half of AI, because inference is thinking, and thinking is far harder than reading.
Last stated 8 months ago
23 Mar 2026
JH
Jensen Huang — holds since 2026-03-23 — tap for who they are
Same subject: Purely conceptual reasoning about AI psychology, without empirical work or hard math, is unlikely to get very far. — tap to centre the map on it
Purely conceptual reasoning about AI psychology, without empirical work or hard math, is unlikely to get very far.
Last stated 4 years ago
12 May 2023
AC
Ajeya Cotra — holds since 2023-05-12 — tap for who they are
Same subject: You need to build early with AI to have any intuition for when the models become capable. — tap to centre the map on it
You need to build early with AI to have any intuition for when the models become capable.
Last stated 2 months ago
10 Sept 2026
MK
Mike Krieger — holds since 2026-09-10 — tap for who they are
Same subject: Building skills takes a long time, and intuition has to be refined. — tap to centre the map on it
Building skills takes a long time, and intuition has to be refined.
Last stated 10 months ago
18 Jan 2026
DS
David Senra — holds since 2026-01-18 — tap for who they are
Same subject: Teaching people a decision model is useless, because the intuition it rests on is built from experience and has no shortcut. — tap to centre the map on it
Teaching people a decision model is useless, because the intuition it rests on is built from experience and has no shortcut.
Last stated 4 years ago
9 Aug 2022
GK
Gary Klein — holds since 2022-08-09 — tap for who they are
Same subject: If humans have access to inspiration, it makes no sense to treat machines as categorically barred from intuition. — tap to centre the map on it
If humans have access to inspiration, it makes no sense to treat machines as categorically barred from intuition.
Last stated 3 weeks ago
31 Oct 2026
ML
Michael Levin — holds since 2026-10-31 — tap for who they are
Same subject: AI excels at certain tasks but cannot substitute for human thinking. — tap to centre the map on it
AI excels at certain tasks but cannot substitute for human thinking.
Last stated 3 months ago
1 Sept 2026
NM
Nick Maggiulli — holds since 2026-09-01 — tap for who they are
Same subject: AI-written code is fine when only the visual result matters, but insufficient in technical domains where understanding what is actually being computed is required. — tap to centre the map on it
AI-written code is fine when only the visual result matters, but insufficient in technical domains where understanding what is actually being computed is required.
Last stated 5 months ago
16 Jun 2026
SW
Stephen Wolfram — holds since 2026-06-16 — 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 not computer vision but the discovery of a universal function approximator that can learn any function.
Last stated 4 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 unify pixel generation and pixel reconstruction, two tracks computer vision has kept apart for over half a century.
Last stated 3 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 are specifically designed to handle variability in two-dimensional shapes.
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 allow multimodule document recognition systems to be trained globally with gradient methods to minimize overall performance.
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, not any particular architecture trick, is what most visual recognition tasks 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 on unfiltered, internet-scale image data 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: Today's AI cannot count the chairs in a video of a few office rooms, something a toddler or young child could do. — tap to centre the map on it
Today's AI cannot count the chairs in a video of a few office rooms, something a toddler or young child could do.
Last stated a year ago
16 Nov 2025
FL
Fei-Fei Li — holds since 2025-11-16 — tap for who they are
Same subject: A program that generates rich enough tests of intelligence would be as hard to build as AGI itself. — tap to centre the map on it
A program that generates rich enough tests of intelligence would be as hard to build as AGI itself.
Last stated 2 years ago
6 Nov 2024
FC
François Chollet — holds since 2024-11-06 — tap for who they are
Same subject: A suitably advanced digital computer can acquire general intelligence. — tap to centre the map on it
A suitably advanced digital computer can acquire general intelligence.
Last stated 12 years ago
16 Jan 2015
SH
Sam Harris — holds since 2015-01-16 — tap for who they are
same subject or similar wording a cloud: claims about one subject, named for it bar: when it was last stated, on a scale from 1998 to today (stretched back to the oldest claim here) — full is today a face: someone who holds the claim — tap it for who they are
At the centre
Visual intuition is probably harder to train AI for than purely symbolic reasoning.
Last stated 18 Sept 2026 · 2 months ago
Holds Grant Sanderson
Read this korrent →
Similar wording
Inference was never going to be the easy, cheap half of AI, because inference is thinking, and thinking is far harder than reading.
Last stated 23 Mar 2026 · 8 months ago
Holds Jensen Huang
Similar wording
Purely conceptual reasoning about AI psychology, without empirical work or hard math, is unlikely to get very far.
Last stated 12 May 2023 · 4 years ago
Holds Ajeya Cotra
Similar wording
You need to build early with AI to have any intuition for when the models become capable.
Last stated 10 Sept 2026 · 2 months ago
Holds Mike Krieger
Similar wording
Building skills takes a long time, and intuition has to be refined.
Last stated 18 Jan 2026 · 10 months ago
Holds David Senra
Similar wording
Teaching people a decision model is useless, because the intuition it rests on is built from experience and has no shortcut.
Last stated 9 Aug 2022 · 4 years ago
Holds Gary Klein
Similar wording
If humans have access to inspiration, it makes no sense to treat machines as categorically barred from intuition.
Last stated 31 Oct 2026 · 3 weeks ago
Holds Michael Levin
Similar wording
AI excels at certain tasks but cannot substitute for human thinking.
Last stated 1 Sept 2026 · 3 months ago
Holds Nick Maggiulli
Similar wording
AI-written code is fine when only the visual result matters, but insufficient in technical domains where understanding what is actually being computed is required.
Last stated 16 Jun 2026 · 5 months ago
Holds Stephen Wolfram
Same subject: computer vision
AlexNet's real breakthrough was not computer vision but the discovery of a universal function approximator that can learn any function.
Last stated 26 Jul 2026 · 4 months ago
Holds Jensen Huang
Same subject: computer vision
Atlas is the first model to unify pixel generation and pixel reconstruction, two tracks computer vision has kept apart for over half a century.
Last stated 4 Sept 2026 · 3 months ago
Holds Fei-Fei Li
Same subject: computer vision
Convolutional neural networks are specifically designed to handle variability in two-dimensional shapes.
Last stated 1 Jan 1998 · 29 years ago
Holds Yoshua Bengio
Same subject: neural networks
Graph transformer networks allow multimodule document recognition systems to be trained globally with gradient methods to minimize overall performance.
Last stated 1 Jan 1998 · 29 years ago
Holds Yoshua Bengio
Same subject: neural networks
The depth of a representation, not any particular architecture trick, is what most visual recognition tasks turn on.
Last stated 10 Dec 2015 · 11 years ago
Holds Jian Sun Kaiming He
Same subject: neural networks
Unsupervised learning trained on unfiltered, internet-scale image data now genuinely works.
Last stated 6 Apr 2021 · 6 years ago
Holds Gwern
Same subject: measuring intelligence
Today's AI cannot count the chairs in a video of a few office rooms, something a toddler or young child could do.
Last stated 16 Nov 2025 · a year ago
Holds Fei-Fei Li
Same subject: measuring intelligence
A program that generates rich enough tests of intelligence would be as hard to build as AGI itself.
Last stated 6 Nov 2024 · 2 years ago
Holds François Chollet
Same subject: measuring intelligence
A suitably advanced digital computer can acquire general intelligence.
Last stated 16 Jan 2015 · 12 years ago
Holds Sam Harris