I feel like I'm the only one not getting the world models hype. We've been talking about them for decades now, and all of it is still theoretical. Meanwhile LLMs and text foundation models showed up, proved to be insanely effective, took over the industry, and people are still going "nah LLMs aren't it, world models will be the gold standard, just wait."
I bet LLMs and world models will merge. World models essentially try to predict the future, with or without actions taken. LLMs with tokenized image input can also be made to predict the future image tokens. It's a very valuable supervised learning signal aside from pre-training and various forms of RL.
I think "world models" is the wrong thing to focus on when contrasting the "animal intelligence" approach (which is what LeCun is striving for) with LLMs, especially since "world model" means different things to different people. Some people would call the internal abstractions/representations that an LLM learns during training a "world model" (of sorts).
The fundamental problem with today's LLMs that will prevent them from achieving human level intelligence, and creativity, is that they are trained to predict training set continuations, which creates two very major limitations:
1) They are fundamentally a COPYING technology, not a learning or creative one. Of course, as we can see, copying in this fashion will get you an extremely long way, especially since it's deep patterns (not surface level text) being copied and recombined in novel ways. But, not all the way to AGI.
2) They are not grounded, therefore they are going to hallucinate.
The animal intelligence approach, the path to AGI, is also predictive, but what you predict is the external world, the future, not training set continuations. When your predictions are wrong (per perceptual feedback) you take this as a learning signal to update your predictions to do better next time a similar situation arises. This is fundamentally a LEARNING architecture, not a COPYING one. You are learning about the real world, not auto-regressively copying the actions that someone else took (training set continuations).
Since the animal is also acting in the external world that it is predicting, and learning about, this means that it is learning the external effects of it's own actions, i.e. it is learning how to DO things - how to achieve given outcomes. When put together with reasoning/planning, this allows it to plan a sequence of actions that should achieve a given external result ("goal").
Since the animal is predicting the real world, based on perceptual inputs from the real world, this means that it's predictions are grounded in reality, which is necessary to prevent hallucinations.
So, to come back to "world models", yes an animal intelligence/AGI built this way will learn a model of how the world works - how it evolves, and how it reacts (how to control it), but this behavioral model has little in common with the internal generative abstractions that an LLM will have learnt, and it is confusing to use the same name "world model" to refer to them both.
RL on LLMs has changed things. LLMs are not stuck in continuation predicting territory any more.
Models build up this big knowledge base by predicting continuations. But then their RL stage gives rewards for completing problems successfully. This requires learning and generalisation to do well, and indeed RL marked a turning point in LLM performance.
A year after RL was made to work, LLMs can now operate in agent harnesses over 100s of tool calls to complete non-trivial tasks. They can recover from their own mistakes. They can write 1000s of lines of code that works. I think it’s no longer fair to categorise LLMs as just continuation-predictors.
Thanks for saying this. It never ceases to amaze me how many people still talk about LLMs like it’s 2023, completely ignoring the RLVR revolution that gave us models like Opus that can one-shot huge chunks of works-first-time code for novel use cases. Modern LLMs aren’t just trained to guess the next token, they are trained to solve tasks.
> The tech industry often talks about “the cloud” as though it were something abstract and untouchable. But the cloud runs on data centers, those data centers have an address, and that address can be hit by a drone.
Nominating this as the best opening line I have read in a while.
> The missing part is that current gpus are already money making machine in 2026
Are they? Unless you are Nvidia that is very far from the case.
OpenAI's current revenue is $25 billion a year. They are expected to spend $600 billion on infrastructure in the next 4 years to sustain and grow that revenue.
Amazon, Google, Microsoft and Meta are spending a combined $650 billion on infrastructure in 2026 alone.
The story is the same across the rest of the industry.
None of these investments are immediately profitable. And it remains to be seen whether they eventually will be or not.
Anthropic in 2026 only added several billions of revenue. This is insanely fast. In my company llm cost are already eating hiring budgets to a certain extent. We don’t buy gpu. We are paying to those who will.
25 bln from just one company . There will be 6-7 companies like this . And they just scratched the surface . The penetration in many areas is almost 0. Yet.
Until there is a capable open source open weight AI that is easily hostable by an average person - no, we still have a long way to go. You aren't going to have software freedom when the tool that enables it is controlled by a handful of powerful tech companies.
reply