Yann LeCun raises $1B to build AI that understands the physical world • Hacker News Board

Re: Yann LeCun raises $1B to build AI that understands the physical world

#351

Post by the_mar » Wed, Mar 11, 2026, 3:18 AM UTC

edanm wrote: post #291

lich_king wrote: post #243

Earlier quoted context omitted.

> The Wason selection task is the classic example: most people fail a simple conditional reasoning problem unless it’s dressed up in familiar social context, like catching cheaters. I've never heard about the Wason selection task, looked it up, and could tell the right answer right away. But I can also tell you why: because I have some familiarity with formal logic and can, in your words, pattern-match the gotcha tha…

Agree with much of your comment. Though note that as GP said, on the Wason selection task, people famously do much better when it's framed in a social context. That at least partially undermines your theory that its lack of familiarity with the terminology of formal logic.

I for the life of me could not solve the <18 example from wikipedia. but the number/color one is super easy

Re: Yann LeCun raises $1B to build AI that understands the physical world

#352

Post by latentsea » Wed, Mar 11, 2026, 3:18 AM UTC

andy12_ wrote: post #62

zelphirkalt wrote: post #37

Earlier quoted context omitted.

> I don't understand this view. How I see it the fundamental bottleneck to AGI is continual learning and backpropagation. Models today are static, and human brains don't learn or adapt themselves with anything close to backpropagation. Even with continuous backpropagation and "learning", enriching the training data, so called online-learning, the limitations will not disappear. The LLMs will not be able to conclude t…

> Even with continuous backpropagation and "learning" That's what I said. Backpropagation cannot be enough; that's not how neurons work in the slightest. When you put biological neurons in a Pong environment they learn to play not through some kind of loss or reward function; they self-organize to avoid unpredictable stimulation. As far as I know, no architecture learns in such an unsupervised way. https://www.scienc…

So can't you just use how real neurons learn as training data to to learn how to learn the same way?

Re: Yann LeCun raises $1B to build AI that understands the physical world

#353

Post by crystal_revenge » Wed, Mar 11, 2026, 3:20 AM UTC

sofixa wrote: post #287

crystal_revenge wrote: post #219

Earlier quoted context omitted.

Europe doesn't want to be reliant (understandably) on the US military for defense, because if they are, as Trump has demonstrated, they will be pressured to make concessions not in their interests. The need for a military is tightly coupled with the EU's need for energy. You can see this in the immediate impact that the war in Iran has had on Germany's natural gas prices [0]. But already unable to defend itself from…

> But already unable to defend itself from Russia, EU countries I'm sorry, but this is just crazy talk. Russia cannot enforce its will on Ukraine, one of the poorest and most corrupt countries in Europe, with a (at time of invasion) relatively small and underequipped army. Yes it has grown through conscription, has been equipped by foreign and domestic supplies, has made some brilliant advances in tech and tactics...…

It’s funny how you basically explain precisely why the war in Ukraine has gone on so long but refuse to recognize it.

As Russia’s economy has continually reshaped over the last 4 years there has been increasingly a domestic demand for war. You point out all the evidence yourself:

> Yes it has grown through conscription, has been equipped by foreign and domestic supplies, has made some brilliant advances in tech and tactics...

Russia (well its oligarchs and rulers) has increasingly benefited from perpetual war. Yes, soon it will need to switch positions to expansion to maintain its economy, but this situation in Iran presents a perfect opportunity if things play it Russia’s interests.

You also will find that if you paid any attention to European politics over the years this is a serious topic to all leaders there.

But I don’t mind if you’re not convinced, I had similar people on hacker news unconvinced Russia could sustain operations in Russia longer than a few months because they were doing so poorly… 4 years ago.

Re: Yann LeCun raises $1B to build AI that understands the physical world

#354

Post by anon7000 » Wed, Mar 11, 2026, 3:32 AM UTC

andy12_ wrote: post #30

A_D_E_P_T wrote: post #11

Justifiable. There are a lot more degrees of freedom in world models. LLMs are fundamentally capped because they only learn from static text -- human communications about the world -- rather than from the world itself, which is why they can remix existing ideas but find it all but impossible to produce genuinely novel discoveries or inventions. A well-funded and well-run startup building physical world models (ground…

I don't understand this view. How I see it the fundamental bottleneck to AGI is continual learning and backpropagation. Models today are static, and human brains don't learn or adapt themselves with anything close to backpropagation. World models don't solve any of these problems; they are fundamentally the same kind of deep learning architectures we are used to work with. Heck, if you think learning from the world i…

I don’t understand your view. Reality is that we need some way to encode the rules of the world in a more definitive way. If we want models to be able to make assertive claims about important information and be correct, it’s very fair to theorize they might need a more deterministic approach than just training them more. But it’s just a theory that this will actually solve the problem.

Ultimately, we still have a lot to learn and a lot of experiments to do. It’s frankly unscientific to suggest any approaches are off the table, unless the data & research truly proves that. Why shouldn’t we take this awesome LLM technology and bring in more techniques to make it better?

A really, really basic example is chess. Current top AI models still don’t know how to play it (https://www.software7.com/blog/ai_chess_vs_1983_atari/) The models are surely trained on source material that include chess rules, and even high level chess games. But the models are not learning how to play chess correctly. They don’t have a model to understand how chess actually works — they only have a non-deterministic prediction based on what they’ve seen, even after being trained on more data than any chess novice has ever seen about the topic. And this is probably one of the easiest things for AI to stimulate. Very clear/brief rules, small problem space, no hidden information, but it can’t handle the massive decision space because its prediction isn’t based on the actual rules, but just “things that look similar”

(And yeah, I’m sure someone could build a specific LLM or agent system that can handle chess, but the point is that the powerful general purpose models can’t do it out of the box after training.)

Maybe more training & self-learning can solve this, but it’s clearly still unsolved. So we should definitely be experimenting with more techniques.

Re: Yann LeCun raises $1B to build AI that understands the physical world

#355

Post by echelon » Wed, Mar 11, 2026, 3:47 AM UTC

chriskanan wrote: post #284

I had lunch with Yann last August, about a week after Alex Wang became his "boss." I asked him how he felt about that, and at the time he told me he would give it a month or two and see how it goes, and then figure out if he should stay or find employment elsewhere. I told him he ought to just create his own company if he decides to leave Meta to chase his own dream, rather than work on the dream's of others. That sa…

You have to understand the strategy of all the other players:

Build attention-grabbing, monetizable models that subsidize (at least in part) the run up to AGI.

Nobody is trying to one-shot AGI. They're grinding and leveling up while (1) developing core competencies around every aspect of the problem domain and (2) winning users.

I don't know if Meta is doing a good job of this, but Google, Anthropic, and OpenAI are.

Trying to go straight for the goal is risky. If the first results aren't economically viable or extremely exciting, the lab risks falling apart.

This is the exact point that Musk was publicly attacking Yann on, and it's likely the same one that Zuck pressed.

Re: Yann LeCun raises $1B to build AI that understands the physical world

#356

Post by programjames » Wed, Mar 11, 2026, 3:47 AM UTC

hdivider wrote: post #347

It's curious to me why we have no theory of intelligence. By which I mean an actual hard and verified theory, as in physics for gravity, electromagnetism, quantum mechanics. Intelligence is simply not well-understood at a mathematical level. Like medieval engineers, we rely so heavily on experimentation in AI. We have no idea how far away from the human level we actually are. Or how far above the human level we can g…

I think this is the equivalent of a non-nuclear physicist asking, "why do we have no theory of nuclear physics?" in the late 1930s. Some people do, they're just not sharing it.

Re: Yann LeCun raises $1B to build AI that understands the physical world

#357

Post by aurareturn » Wed, Mar 11, 2026, 4:09 AM UTC

jacquesm wrote: post #311

andy12_ wrote: post #30

Earlier quoted context omitted.

I don't understand this view. How I see it the fundamental bottleneck to AGI is continual learning and backpropagation. Models today are static, and human brains don't learn or adapt themselves with anything close to backpropagation. World models don't solve any of these problems; they are fundamentally the same kind of deep learning architectures we are used to work with. Heck, if you think learning from the world i…

The main difference is humans are learning all the time and models learn batch wise and forget whatever happened in a previous session unless someone makes it part of the training data so there is a massive lag. Whoever cracks the continuous customized (per user, for instance) learning problem without just extending the context window is going to be making a big splash. And I don't mean cheats and shortcuts, I mean a…

Why not just provide more compute for say, 1 billion token context for each user to mimic continuous learning. Then retrain the model in the background to include learnings.

The user wouldn’t know if the continuous learning came from the context or the model retrained. It wouldn’t matter.

Continuous learning seems to be a compute and engineering problem.

Re: Yann LeCun raises $1B to build AI that understands the physical world

#358

Post by tellarin » Wed, Mar 11, 2026, 4:11 AM UTC

Selfless plug here... Some collaborators and I just released a first version of a benchmark we think highlights a critical gap in recent models in understanding causality in the real-world, beyond a physics focus.

Everyday environments are rich in tangible control interfaces (TCIs), like, light switches, appliance panels, and embedded GUIs, that are designed for humans and demand commonsense and physics reasoning, but also causal prediction and outcome verification in time and space (e.g., delayed heating, remote lights).

SWITCH: Benchmarking Modeling and Handling of Tangible Interfaces in Long-horizon Embodied Scenarios (https://huggingface.co/papers/2511.17649)

Feedback, suggestions, and collaborators are very welcome!

Re: Yann LeCun raises $1B to build AI that understands the physical world

#359

Post by nextaccountic » Wed, Mar 11, 2026, 4:34 AM UTC

perfmode wrote: post #232

zelphirkalt wrote: post #37

Earlier quoted context omitted.

> I don't understand this view. How I see it the fundamental bottleneck to AGI is continual learning and backpropagation. Models today are static, and human brains don't learn or adapt themselves with anything close to backpropagation. Even with continuous backpropagation and "learning", enriching the training data, so called online-learning, the limitations will not disappear. The LLMs will not be able to conclude t…

Humans are notoriously bad at formal logic. The Wason selection task is the classic example: most people fail a simple conditional reasoning problem unless it’s dressed up in familiar social context, like catching cheaters. That looks a lot more like pattern matching than rule application. Kahneman’s whole framework points the same direction. Most of what people call “reasoning” is fast, associative, pattern-based. T…

> Kahneman’s whole framework points the same direction. Most of what people call “reasoning” is fast, associative, pattern-based. The slow, deliberate, step-by-step stuff is effortful and error-prone, and people avoid it when they can. And even when they do engage it, they’re often confabulating a logical-sounding justification for a conclusion they already reached by other means.

Some references on that

https://en.wikipedia.org/wiki/Thinking,_Fast_and_Slow

https://thedecisionlab.com/reference-guide/philosophy/system...

System 1 really looks like a LLM (indeed completing a phrase is an example of what it can do, like, "you either die a hero, or you live enough to become the _"). It's largely unconscious and runs all the time, pattern matching on random stuff

System 2 is something else and looks like a supervisor system, a higher level stuff that can be consciously directed through your own will

But the two systems run at the same time and reinforce each other

Re: Yann LeCun raises $1B to build AI that understands the physical world

#360

Post by nextaccountic » Wed, Mar 11, 2026, 5:00 AM UTC

lich_king wrote: post #243

perfmode wrote: post #232

Earlier quoted context omitted.

Humans are notoriously bad at formal logic. The Wason selection task is the classic example: most people fail a simple conditional reasoning problem unless it’s dressed up in familiar social context, like catching cheaters. That looks a lot more like pattern matching than rule application. Kahneman’s whole framework points the same direction. Most of what people call “reasoning” is fast, associative, pattern-based. T…

> The Wason selection task is the classic example: most people fail a simple conditional reasoning problem unless it’s dressed up in familiar social context, like catching cheaters. I've never heard about the Wason selection task, looked it up, and could tell the right answer right away. But I can also tell you why: because I have some familiarity with formal logic and can, in your words, pattern-match the gotcha tha…

> If you added a simple additional to the problem, such as "Note that in this context, 'if' only means that...", most people would almost certainly answer it correctly.

Agreed. More broadly, classical logic isn't the only logic out there. Many logics will differ on the meaning of implication if x then y. There's multiple ways for x to imply y, and those additional meanings do show up in natural language all the time, and we actually do have logical systems to describe them, they are just lesser known.

Mapping natural language into logic often requires a context that lies outside the words that were written or spoken. We need to represent into formulas what people actually meant, rather than just what they wrote. Indeed the same sentence can be sometimes ambiguous, and a logical formula never is.

As an aside, I wanna say that material implication (that is, the "if x then y" of classical logic) deeply sucks, or rather, an implication in natural language very rarely maps cleanly into material implication. Having an implication if x then y being vacuously true when x is false is something usually associated with people that smirk on clever wordplays, rather than something people actually mean when they say "if x then y"