Tap a claim on the ring to put it at the centre.
← The costs of building true frontier models are now measured in…
17 connected korrents · 14 moments from 12 Jun 2017 to 17 Oct 2026.
Everything filed under LLMs
LLMs
Everything filed under scaling laws
scaling laws
Everything filed under OpenAI
OpenAI
Everything filed under transformers
transformers
Everything filed under innovation
innovation
Everything filed under Anthropic
Anthropic 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 costs of building true frontier models are now measured in billions per major model, meaning only huge companies can sustain them. The costs of building true frontier models are now measured in billions per major model, meaning only huge companies can sustain them. Last stated today
17 Oct 2026
NL
Nathan Lambert — holds since 2026-10-17 — tap for who they are
Same subject: Every frontier model is profitable on its own; the company loses money only because it is already training one ten times more expensive. — tap to centre the map on it Every frontier model is profitable on its own; the company loses money only because it is already training one ten times more expensive. Last stated 8 months ago
13 Feb 2026
DA
Dario Amodei — holds since 2026-02-13 — tap for who they are
Same subject: Frontier models must be cheap enough for anyone to use, because a world in which they are rationed to the wealthiest developers is the wrong world. — tap to centre the map on it Frontier models must be cheap enough for anyone to use, because a world in which they are rationed to the wealthiest developers is the wrong world. Last stated 3 months ago
29 Jul 2026
AW
Alexandr Wang — holds since 2026-07-29 — tap for who they are
Same subject: The model companies' margins will rise this year, because being capacity-constrained forces them to destroy their own demand. — tap to centre the map on it The model companies' margins will rise this year, because being capacity-constrained forces them to destroy their own demand. Last stated 7 months ago
13 Mar 2026
DP
Dylan Patel — holds since 2026-03-13 — tap for who they are
Same subject: A competitor six months or a year behind the frontier can stay close, because the cost of AI falls more than tenfold every year. — tap to centre the map on it A competitor six months or a year behind the frontier can stay close, because the cost of AI falls more than tenfold every year. Last stated 9 months ago
8 Jan 2026
JH
Jensen Huang — holds since 2026-01-08 — tap for who they are
Same subject: Most future growth in large language models' practical value will come from how they are harnessed and connected to other tools, not from the core models alone. — tap to centre the map on it Most future growth in large language models' practical value will come from how they are harnessed and connected to other tools, not from the core models alone. Last stated 8 months ago
23 Feb 2026
SW
Stephen Wolfram — holds since 2026-02-23 — tap for who they are
Same subject: Companies with sunk cost in an existing paradigm find it very painful to build the new thing from scratch. — tap to centre the map on it Companies with sunk cost in an existing paradigm find it very painful to build the new thing from scratch. Last stated a month ago
8 Sept 2026
RU
Roman Ugarte — holds since 2026-09-08 — tap for who they are
Same subject: Open-source models cannot keep frontier systems in check, because they will always be much dumber than the frontier. — tap to centre the map on it Open-source models cannot keep frontier systems in check, because they will always be much dumber than the frontier. Last stated 2 months ago
1 Sept 2026
AC
Ajeya Cotra — holds since 2026-09-01 — tap for who they are
Same subject: Local AI models will eventually catch up to frontier models once frontier progress hits diminishing returns. — tap to centre the map on it Local AI models will eventually catch up to frontier models once frontier progress hits diminishing returns. Last stated 8 months ago
8 Feb 2026
DC
David Crawshaw — holds since 2026-02-08 — tap for who they are
Same subject: A startup should not begin life as a nonprofit and bolt a for-profit arm on later, whatever OpenAI's own history suggests. — tap to centre the map on it A startup should not begin life as a nonprofit and bolt a for-profit arm on later, whatever OpenAI's own history suggests. Last stated 3 years ago
18 Mar 2024
SA
Sam Altman — holds since 2024-03-18 — tap for who they are
Same subject: Advertising was a necessary phase for the internet but a momentary industry, and an AI people pay for is better because they know the answers are not influenced by advertisers. — tap to centre the map on it Advertising was a necessary phase for the internet but a momentary industry, and an AI people pay for is better because they know the answers are not influenced by advertisers. Last stated 3 years ago
18 Mar 2024
SA
Sam Altman — holds since 2024-03-18 — tap for who they are
Same subject: Agent containment will be one of the most important priorities for any AI infra company. — tap to centre the map on it Agent containment will be one of the most important priorities for any AI infra company. Last stated 3 days ago
14 Oct 2026
GR
Guillermo Rauch — holds since 2026-10-14 — 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: A transduction model can rely entirely on self-attention for input and output representations without sequence-aligned RNNs or convolution. — tap to centre the map on it A transduction model can rely entirely on self-attention for input and output representations without sequence-aligned RNNs or convolution. Last stated 9 years ago
12 Jun 2017
IP
Illia Polosukhin — holds since 2017-06-12 — tap for who they are
JU
Jakob Uszkoreit — holds since 2017-06-12 — tap for who they are
NP
Niki Parmar — holds since 2017-06-12 — tap for who they are
AV
Ashish Vaswani — holds since 2017-06-12 — tap for who they are
NS
Noam Shazeer — holds since 2017-06-12 — tap for who they are
AG
Aidan N. Gomez — holds since 2017-06-12 — tap for who they are
+2
2 more
Same subject: Self-attention can yield more interpretable models. — tap to centre the map on it Self-attention can yield more interpretable models. Last stated 9 years ago
12 Jun 2017
IP
Illia Polosukhin — holds since 2017-06-12 — tap for who they are
JU
Jakob Uszkoreit — holds since 2017-06-12 — tap for who they are
NP
Niki Parmar — holds since 2017-06-12 — tap for who they are
AV
Ashish Vaswani — holds since 2017-06-12 — tap for who they are
NS
Noam Shazeer — holds since 2017-06-12 — tap for who they are
AG
Aidan N. Gomez — holds since 2017-06-12 — tap for who they are
+2
2 more
Same subject: Sequence transduction can use a simple architecture based solely on attention, without recurrence or convolutions. — tap to centre the map on it Sequence transduction can use a simple architecture based solely on attention, without recurrence or convolutions. Last stated 9 years ago
12 Jun 2017
IP
Illia Polosukhin — holds since 2017-06-12 — tap for who they are
JU
Jakob Uszkoreit — holds since 2017-06-12 — tap for who they are
NP
Niki Parmar — holds since 2017-06-12 — tap for who they are
AV
Ashish Vaswani — holds since 2017-06-12 — tap for who they are
NS
Noam Shazeer — holds since 2017-06-12 — tap for who they are
AG
Aidan N. Gomez — holds since 2017-06-12 — tap for who they are
+2
2 more
At the centre
The costs of building true frontier models are now measured in billions per major model, meaning only huge companies can sustain them.
Holds Read this korrent →
Every frontier model is profitable on its own; the company loses money only because it is already training one ten times more expensive.
Holds
Frontier models must be cheap enough for anyone to use, because a world in which they are rationed to the wealthiest developers is the wrong world.
Holds
The model companies' margins will rise this year, because being capacity-constrained forces them to destroy their own demand.
Holds
A competitor six months or a year behind the frontier can stay close, because the cost of AI falls more than tenfold every year.
Holds
Most future growth in large language models' practical value will come from how they are harnessed and connected to other tools, not from the core models alone.
Holds
Companies with sunk cost in an existing paradigm find it very painful to build the new thing from scratch.
Holds
Open-source models cannot keep frontier systems in check, because they will always be much dumber than the frontier.
Holds
Local AI models will eventually catch up to frontier models once frontier progress hits diminishing returns.
Holds
A startup should not begin life as a nonprofit and bolt a for-profit arm on later, whatever OpenAI's own history suggests.
Holds
Advertising was a necessary phase for the internet but a momentary industry, and an AI people pay for is better because they know the answers are not influenced by advertisers.
Holds
Agent containment will be one of the most important priorities for any AI infra company.
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
A transduction model can rely entirely on self-attention for input and output representations without sequence-aligned RNNs or convolution.
Holds
Self-attention can yield more interpretable models.
Holds
Sequence transduction can use a simple architecture based solely on attention, without recurrence or convolutions.
Holds