Yann LeCun
Professor of Computer Science at New York University and 2018 Turing Award laureate; a leading researcher in deep learning, known for convolutional neural networks.
Yann LeCun did not write this page. ppll compiled it from their own words, each linked to the source. Is this you? · Report a mistake
Latest x.com
What they believe
-
Auto-regressive LLMs in their current form will not take us to human-level AI
So no, Auto-Regressive LLMs in their current form will not take us to human-level AI.
x.comSaid 31 Aug 2024This on korrents
-
Regulating AI deployment is fine, but regulating AI research and development obstructs open research and open-source AI
SB1047 attempts to regulate AI research and development, creating obstacles to the dissemination open research in AI and open source AI platforms. Regulating the deployment of AI applications is fine.
x.comSaid 28 Jun 2024This on korrents
The bold line is korrents’ wording; the quote is theirs. These 2: what they wrote before what they said aloud, one per source, newest first.
Feed
As its own page →Hiding
7 November
15 December 2024
2 December 2024
29 November 2024
28 November 2024
27 November 2024
24 November 2024
Nothing matches.
Beliefs
Korrents What they believe 8 beliefs — each backed by an exact quote.
Each is a — compiled by korrents.com, not by them: the one-line wordings are korrents', the quotes are theirs.
Recent
Auto-regressive LLMs in their current form will not take us to human-level AI
So no, Auto-Regressive LLMs in their current form will not take us to human-level AI.
@ylecun on X Said 31 Aug 2024
Regulating AI deployment is fine, but regulating AI research and development obstructs open research and open-source AI
SB1047 attempts to regulate AI research and development, creating obstacles to the dissemination open research in AI and open source AI platforms. Regulating the deployment of AI applications is fine.
@ylecun on X Said 28 Jun 2024
Auto-regressive LLMs that spend a fixed amount of computation per token cannot reason, whatever the architecture
The problem isn't that it is a transformer. The problem is that it is an auto-regressive LLM. Auto-regressive LLMs that compute each token with a fixed number of computational steps can't reason, regardless of the details of the architecture.
@ylecun on X Said 23 Jun 2024
Show 5 more
Regulators should regulate AI applications, not the underlying technology
Regulators should regulate applications, not technology. - Regulating basic technology will put an end to innovation.
@ylecun on X Said 6 Jun 2024
Open-source AI models are not more dangerous than closed ones
A big bunch of nails in the coffin of the idea that open source AI models are more dangerous than closed ones.
@ylecun on X Said 19 Jan 2024
Scaling up auto-regressive LLMs will not fix their very limited reasoning and planning abilities
AR-LLMs have very limited reasoning and planning abilities. This will not be fixed by making them bigger and training them on more data.
@ylecun on X Said 23 Oct 2023
Auto-regressive LLMs are useful as writing and coding aids but hallucinate too often
1- Auto-Regressive LLMs are useful, particularly as writing aids, particularly for code. 2- they hallucinate too often
@ylecun on X Said 26 Mar 2023
LLMs write code better than general text because a program's universe is limited, discrete, deterministic and fully observable, unlike the real world
Because, unlike the real world, the universe that a program manipulates (the state of the variables) is limited, discrete, deterministic, and fully observable. The real world is none of that.
@ylecun on X Said 13 Feb 2023
What is a korrent?
A korrent is a belief a person has stated in their own words: one sentence stating the claim, backed by a quote and a source, kept at korrents.com.
Under a name here, the quoted block is what they actually said. The korrent beneath it is the claim those words support, in korrents' wording — tap it to see the record, its source, and who else holds it.
Nobody here wrote their own korrents. They are compiled from public statements, and a person can change their mind, which is recorded too.
Where they publish
x.com @ylecun Personal account.
Recent
- Reminder from 3 years ago. 7 Nov 2026
Link verified 25 Sept 2026. Recent items update automatically from the channel.
Bluesky yann-lecun.bsky.social Personal account; handle carries his own name.
Recent
- Aravind Srinivas (Perplexity) tells Lex all the ways in which I was right against the prevalent ideas of the time: DL, ConvNets, energy-based models, SSL, the limitation of RL, and now the limitations of auto-regressive generative models including LLMs. Thanks Aravind! 15 Dec 2024 youtu.be
- The total distance flown passengers on US airlines since the last crash is 2.3 light-years. That's about 2.2*10^13 km. The last crash was on Feb 12, 2009. 2 Dec 2024 ourworldindata.org
- Another thing to be thankful for: the use of AI to help improve climate change models. From The Economist: www.economist.com/science-and-... Free access on archive.org: 29 Nov 2024 archive.is
Show 6 more
- Excellent podcast with Nikhil Kamath in which we cover a lot of topics related to AI and deep learning. 29 Nov 2024 youtube.com
- WaPo Edito: Be thankful for the applications of AI in medicine. More accurate detection of cancers (breast, prostate, skin, brain), faster diagnosis of strokes, sepsis, heart attacks, faster MRIs, full-body in 40 minutes. Much more to come over the next years. 28 Nov 2024 washingtonpost.com
- Video of my chat with Gaurav Agarwal on the future of AI, world models, planning/reasoning, the limitations of LLMs, AI as an empirical science, why neural net were shunned. And how countries could jumpstart their AI ecosystem by creating top industry research labs. 27 Nov 2024 youtu.be
- Auto-Regressive LLMs (auto-encoders with causal transformer architectures) and BERT-style models (denoising auto-encoders with transformer archis) are smashing demonstrations of the power of self-supervised (pre-)training. But they only work for sequences of discrete symbols: language, proteins... 24 Nov 2024
- I first showed this "cake" slide in 2016. 24 Nov 2024
- @ylecun.bsky.social Hi there, bsky dwellers! Here is Messier 51, AKA the Whirlpool Galaxy. Shot from a New Jersey backyard with a Celestron EdgeHD 11", 0.7 reducer, ASI6200MM monochrome camera, off-axis guider. 131 subs, 300 seconds each. HSO mapped to RGB. 24 Nov 2024
Link verified 25 Sept 2026. Recent items update automatically from the channel.
About the English under a post
Some people here publish in a language other than English. Where they do, this site shows a machine translation beneath the post, in this typeface — the site's own, not theirs.
The post itself is never changed, moved or hidden: what is set in the serif above is exactly what the person published, and it is what to quote them on. A translation can be wrong in ways that matter, especially about tone.
Only the post's own words are translated. A quoted post, a linked article and a belief on korrents.com are left in their original language.