Tap a claim on the ring to put it at the centre.
← The largest tech companies are not using fewer tokens; they are…
17 connected korrents · 17 moments on record from 19 Jun 2024 to 9 Sept 2026.
Everything filed under LLMs
LLMs
Everything filed under scaling laws
scaling laws
Everything filed under code generation
code generation
Everything filed under AI agents
AI agents
Everything filed under data centers
data centers
Everything filed under America
America 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 largest tech companies are not using fewer tokens; they are switching to cheaper, still capable open models instead of SOTA models. The largest tech companies are not using fewer tokens; they are switching to cheaper, still capable open models instead of SOTA models. Last stated 3 days ago
9 Sept 2026
GO
Gergely Orosz — holds since 2026-09-09 — tap for who they are
Same subject: Maxing out your token usage is optimising one node of the factory for utilisation instead of the end-to-end goal of shipping value that is stable and lasts. — tap to centre the map on it Maxing out your token usage is optimising one node of the factory for utilisation instead of the end-to-end goal of shipping value that is stable and lasts. Last stated 2 months ago
15 Jul 2026
DH
Dex Horthy — holds since 2026-07-15 — tap for who they are
Same subject: Token prices have stayed flat because labs deliberately kept models smaller than expected in order to get more experimental cycles. — tap to centre the map on it Token prices have stayed flat because labs deliberately kept models smaller than expected in order to get more experimental cycles. Last stated a month ago
11 Aug 2026
RG
Ryan Greenblatt — holds since 2026-08-11 — tap for who they are
Same subject: The case for sweating every line of code was premised on humans doing the modifications, and it now survives only because tokens are scarce. — tap to centre the map on it The case for sweating every line of code was premised on humans doing the modifications, and it now survives only because tokens are scarce. Last stated 2 weeks ago
26 Aug 2026
DH
David Heinemeier Hansson — holds since 2026-08-26 — tap for who they are
Same subject: The cost of a token is falling by an order of magnitude every year, even as the price of the computers that make them goes up. — tap to centre the map on it The cost of a token is falling by an order of magnitude every year, even as the price of the computers that make them goes up. Last stated 6 months ago
23 Mar 2026
JH
Jensen Huang — holds since 2026-03-23 — tap for who they are
Same subject: Perpetually running proactive agents are not a technology problem but a token problem, and no ordinary subscription would survive one. — tap to centre the map on it Perpetually running proactive agents are not a technology problem but a token problem, and no ordinary subscription would survive one. Last stated a month ago
10 Aug 2026
PS
Peter Steinberger — holds since 2026-08-10 — tap for who they are
Same subject: Hundreds of billions of dollars of AI value rest on about a billion dollars of lithography tooling that cannot be scaled quickly. — tap to centre the map on it Hundreds of billions of dollars of AI value rest on about a billion dollars of lithography tooling that cannot be scaled quickly. Last stated 6 months ago
13 Mar 2026
DP
Dylan Patel — holds since 2026-03-13 — tap for who they are
Same subject: Most of the gain in tokens per dollar per watt comes from software, not from new hardware — 5x, 10x, sometimes 40x on one model family. — tap to centre the map on it Most of the gain in tokens per dollar per watt comes from software, not from new hardware — 5x, 10x, sometimes 40x on one model family. Last stated 10 months ago
12 Nov 2025
SN
Satya Nadella — holds since 2025-11-12 — tap for who they are
Same subject: The worry that AI's returns are locked away in private companies is overstated: well under a fifth of American market capitalisation is private. — tap to centre the map on it The worry that AI's returns are locked away in private companies is overstated: well under a fifth of American market capitalisation is private. Last stated 3 months ago
4 Jun 2026
PT
Phil Trammell — holds since 2026-06-04 — 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 2 months ago
15 Jul 2026
DH
Dex Horthy — holds since 2026-07-15 — tap for who they are
Same subject: A child who has seen ten cats learns what a machine needs the whole internet of cat photos for, by a learning pathway nobody has solved. — tap to centre the map on it A child who has seen ten cats learns what a machine needs the whole internet of cat photos for, by a learning pathway nobody has solved. Last stated a month ago
10 Aug 2026
FL
Fei-Fei Li — holds since 2026-08-10 — tap for who they are
Same subject: A frontier model is measurably more intelligent with no system prompt at all; the prompts that remain are there for the product, not the model. — tap to centre the map on it A frontier model is measurably more intelligent with no system prompt at all; the prompts that remain are there for the product, not the model. Last stated 2 months ago
27 Jul 2026
BC
Boris Cherny — holds since 2026-07-27 — tap for who they are
Same subject: Language models are already a form of AGI; what the labs are chasing is a further step, not the arrival of general intelligence. — tap to centre the map on it Language models are already a form of AGI; what the labs are chasing is a further step, not the arrival of general intelligence. Last stated 2 years ago
3 Feb 2025
NL
Nathan Lambert — holds since 2025-02-03 — tap for who they are
Same subject: The LLM line of research will reach a capability plateau. — tap to centre the map on it The LLM line of research will reach a capability plateau. Last stated a month ago
7 Aug 2026
FC
François Chollet — no longer holds since 2026-08-07 — tap for who they are
GM
Gary Marcus — holds since 2025-06-07 — tap for who they are
Same subject: The memorize-fetch-apply paradigm behind LLMs can reach arbitrary skill given training data, but it cannot adapt to novelty or acquire new skills on the fly. — tap to centre the map on it The memorize-fetch-apply paradigm behind LLMs can reach arbitrary skill given training data, but it cannot adapt to novelty or acquire new skills on the fly. Last stated 2 years ago
20 Dec 2024
FC
François Chollet — holds since 2024-12-20 — tap for who they are
FC
François Chollet — no longer holds since 2024-12-20 — tap for who they are
Same subject: LLMs can write a large fraction of the tedious code a developer will ever need to write, and most code on most projects is tedious. — tap to centre the map on it LLMs can write a large fraction of the tedious code a developer will ever need to write, and most code on most projects is tedious. Last stated a year ago
2 Jun 2025
TP
Thomas Ptacek — holds since 2025-06-02 — tap for who they are
Same subject: Once a language model reads the results, search can trade precision for recall, because the model does not care that the right link came ninth. — tap to centre the map on it Once a language model reads the results, search can trade precision for recall, because the model does not care that the right link came ninth. Last stated 2 years ago
19 Jun 2024
AS
Aravind Srinivas — holds since 2024-06-19 — tap for who they are
Same subject: Proprietary data is not an advantage in training large language models, because no company's holdings are large enough to matter. — tap to centre the map on it Proprietary data is not an advantage in training large language models, because no company's holdings are large enough to matter. Last stated a year ago
2 Sept 2025
BE
Benedict Evans — holds since 2025-09-02 — tap for who they are
At the centre
The largest tech companies are not using fewer tokens; they are switching to cheaper, still capable open models instead of SOTA models.
Holds Read this korrent →
Maxing out your token usage is optimising one node of the factory for utilisation instead of the end-to-end goal of shipping value that is stable and lasts.
Holds DH Dex Horthy
Token prices have stayed flat because labs deliberately kept models smaller than expected in order to get more experimental cycles.
Holds RG Ryan Greenblatt
The case for sweating every line of code was premised on humans doing the modifications, and it now survives only because tokens are scarce.
Holds
The cost of a token is falling by an order of magnitude every year, even as the price of the computers that make them goes up.
Holds JH Jensen Huang
Perpetually running proactive agents are not a technology problem but a token problem, and no ordinary subscription would survive one.
Holds
Hundreds of billions of dollars of AI value rest on about a billion dollars of lithography tooling that cannot be scaled quickly.
Holds DP Dylan Patel
Most of the gain in tokens per dollar per watt comes from software, not from new hardware — 5x, 10x, sometimes 40x on one model family.
Holds SN Satya Nadella
The worry that AI's returns are locked away in private companies is overstated: well under a fifth of American market capitalisation is private.
Holds PT Phil Trammell
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 DH Dex Horthy
A child who has seen ten cats learns what a machine needs the whole internet of cat photos for, by a learning pathway nobody has solved.
Holds FL Fei-Fei Li
A frontier model is measurably more intelligent with no system prompt at all; the prompts that remain are there for the product, not the model.
Holds
Language models are already a form of AGI; what the labs are chasing is a further step, not the arrival of general intelligence.
Holds NL Nathan Lambert
The LLM line of research will reach a capability plateau.
Holds GM Gary MarcusNo longer holds
The memorize-fetch-apply paradigm behind LLMs can reach arbitrary skill given training data, but it cannot adapt to novelty or acquire new skills on the fly.
No longer holds
LLMs can write a large fraction of the tedious code a developer will ever need to write, and most code on most projects is tedious.
Holds
Once a language model reads the results, search can trade precision for recall, because the model does not care that the right link came ninth.
Holds AS Aravind Srinivas
Proprietary data is not an advantage in training large language models, because no company's holdings are large enough to matter.
Holds BE Benedict Evans