Tap a claim on the ring to put it at the centre.
← The companies that already control the largest troves of real-world…
17 connected korrents · 12 moments from 9 Dec 2024 to 4 Sept 2026. Nearly all of them are about world models .
Everything filed under startups
startups
Everything filed under recursive self-improvement
recursive self-improvement
Everything filed under AGI
AGI
Everything filed under scaling laws
scaling laws 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 companies that already control the largest troves of real-world video data will be first to build successful world models. The companies that already control the largest troves of real-world video data will be first to build successful world models. Last stated 3 months ago
10 Aug 2026
KK
Kevin Kelly — holds since 2026-08-10 — tap for who they are
Same subject: An adaptive agent's world model cannot be a static representation learned once and fixed — tap to centre the map on it An adaptive agent's world model cannot be a static representation learned once and fixed Last stated a year ago
17 Jul 2025
FC
François Chollet — holds since 2025-07-17 — tap for who they are
Same subject: Coding is not just another application domain for AI; it is the meta-skill that lets AI generate its own training material and start the recursive self-improvement loop. — tap to centre the map on it Coding is not just another application domain for AI; it is the meta-skill that lets AI generate its own training material and start the recursive self-improvement loop. Last stated 3 months ago
10 Aug 2026
FC
François Chollet — holds since 2026-08-10 — tap for who they are
Same subject: Current AI evaluation of world models is too narrow, favoring static representations learned from massive datasets over efficient learning through interaction — tap to centre the map on it Current AI evaluation of world models is too narrow, favoring static representations learned from massive datasets over efficient learning through interaction Last stated a year ago
17 Jul 2025
FC
François Chollet — holds since 2025-07-17 — tap for who they are
Same subject: Games are particularly rich and controlled environments uniquely well-suited for evaluating rapid world model induction in AI — tap to centre the map on it Games are particularly rich and controlled environments uniquely well-suited for evaluating rapid world model induction in AI Last stated a year ago
17 Jul 2025
FC
François Chollet — holds since 2025-07-17 — tap for who they are
Same subject: Human adaptability is fundamentally linked to the efficient construction and refinement of internal world models — tap to centre the map on it Human adaptability is fundamentally linked to the efficient construction and refinement of internal world models Last stated a year ago
17 Jul 2025
FC
François Chollet — holds since 2025-07-17 — tap for who they are
Same subject: Rapid world model induction will be critical for achieving robust, general AI that functions in complex, fast-changing real and human worlds — tap to centre the map on it Rapid world model induction will be critical for achieving robust, general AI that functions in complex, fast-changing real and human worlds Last stated a year ago
17 Jul 2025
FC
François Chollet — holds since 2025-07-17 — tap for who they are
Same subject: The scaling law for world models was a conviction held in advance, not a discovery; the devils were always in the architecture choices and the data mixture. — tap to centre the map on it The scaling law for world models was a conviction held in advance, not a discovery; the devils were always in the architecture choices and the data mixture. Last stated 2 months ago
4 Sept 2026
FL
Fei-Fei Li — holds since 2026-09-04 — tap for who they are
Same subject: A belief nobody else shares is a warning, and a belief everybody shares is also a warning; the right number of people who agree with you is small but not zero. — tap to centre the map on it A belief nobody else shares is a warning, and a belief everybody shares is also a warning; the right number of people who agree with you is small but not zero. Last stated 3 months ago
28 Jul 2026
SA
Sam Altman — holds since 2026-07-28 — tap for who they are
Same subject: “Why now?” is the wrong question to make a founder answer — the honest answer can simply be that you started now. — tap to centre the map on it “Why now?” is the wrong question to make a founder answer — the honest answer can simply be that you started now. Last stated 3 months ago
28 Jul 2026
BS
Blake Scholl — holds since 2026-07-28 — tap for who they are
Same subject: A big venture firm's edge is deploying capital into what is obviously working; a small firm's is backing a founder or idea that is not yet obvious. — tap to centre the map on it A big venture firm's edge is deploying capital into what is obviously working; a small firm's is backing a founder or idea that is not yet obvious. Last stated a year ago
23 Jul 2025
JS
Jon Sakoda — holds since 2025-07-23 — tap for who they are
Same subject: A billboard buys only short, shallow awareness, so it does not work for education or for a startup before product-market fit. — tap to centre the map on it A billboard buys only short, shallow awareness, so it does not work for education or for a startup before product-market fit. Last stated 5 months ago
9 Jun 2026
EK
Emily Kramer — holds since 2026-06-09 — tap for who they are
Same subject: A first-time founder has an advantage over a repeat winner, because past success makes it too easy to raise money and hire against a big vision before there is product-market fit. — tap to centre the map on it A first-time founder has an advantage over a repeat winner, because past success makes it too easy to raise money and hire against a big vision before there is product-market fit. Last stated 5 months ago
14 Jun 2026
MP
Mark Pincus — holds since 2026-06-14 — tap for who they are
Same subject: A venture capital "regatta", where most of the race happens before the Series A, is replacing the VC factory model. — tap to centre the map on it A venture capital "regatta", where most of the race happens before the Series A, is replacing the VC factory model. Last stated 2 years ago
9 Dec 2024
SL
Sam Lessin — holds since 2024-12-09 — tap for who they are
Same subject: About 5% is the most common final ownership stake for the founder of a VC-funded startup that works out. — tap to centre the map on it About 5% is the most common final ownership stake for the founder of a VC-funded startup that works out. Last stated 4 months ago
18 Jul 2026
PL
Pieter Levels — holds since 2026-07-18 — tap for who they are
Same subject: A founder who does not know their own goal dies by a thousand compromises and ends up running a company they do not want to work at. — tap to centre the map on it A founder who does not know their own goal dies by a thousand compromises and ends up running a company they do not want to work at. Last stated 5 months ago
2 Jun 2026
MP
Mark Pincus — holds since 2026-06-02 — tap for who they are
Same subject: AI research follows a power law, as startup investing does: the best bet outperforms all the others put together. — tap to centre the map on it AI research follows a power law, as startup investing does: the best bet outperforms all the others put together. Last stated 2 months ago
23 Aug 2026
SA
Sam Altman — holds since 2026-08-23 — tap for who they are
Same subject: An investor who passes on a pitch they find truly novel will often still refer it to another investor, to get credit for having found it. — tap to centre the map on it An investor who passes on a pitch they find truly novel will often still refer it to another investor, to get credit for having found it. Last stated 3 months ago
5 Aug 2026
ZH
Zach Holman — holds since 2026-08-05 — tap for who they are
At the centre
The companies that already control the largest troves of real-world video data will be first to build successful world models.
Holds Read this korrent →
An adaptive agent's world model cannot be a static representation learned once and fixed
Holds
Coding is not just another application domain for AI; it is the meta-skill that lets AI generate its own training material and start the recursive self-improvement loop.
Holds
Current AI evaluation of world models is too narrow, favoring static representations learned from massive datasets over efficient learning through interaction
Holds
Games are particularly rich and controlled environments uniquely well-suited for evaluating rapid world model induction in AI
Holds
Human adaptability is fundamentally linked to the efficient construction and refinement of internal world models
Holds
Rapid world model induction will be critical for achieving robust, general AI that functions in complex, fast-changing real and human worlds
Holds
The scaling law for world models was a conviction held in advance, not a discovery; the devils were always in the architecture choices and the data mixture.
Holds
A belief nobody else shares is a warning, and a belief everybody shares is also a warning; the right number of people who agree with you is small but not zero.
Holds
“Why now?” is the wrong question to make a founder answer — the honest answer can simply be that you started now.
Holds
A big venture firm's edge is deploying capital into what is obviously working; a small firm's is backing a founder or idea that is not yet obvious.
Holds
A billboard buys only short, shallow awareness, so it does not work for education or for a startup before product-market fit.
Holds
A first-time founder has an advantage over a repeat winner, because past success makes it too easy to raise money and hire against a big vision before there is product-market fit.
Holds
A venture capital "regatta", where most of the race happens before the Series A, is replacing the VC factory model.
Holds
About 5% is the most common final ownership stake for the founder of a VC-funded startup that works out.
Holds
A founder who does not know their own goal dies by a thousand compromises and ends up running a company they do not want to work at.
Holds
AI research follows a power law, as startup investing does: the best bet outperforms all the others put together.
Holds
An investor who passes on a pitch they find truly novel will often still refer it to another investor, to get credit for having found it.
Holds