Large Language Models or LLMs have taken the world by storm, and are finding applications in fields ranging from programming to finance, but Meta’s AI Chief Yann LeCun feels they might’ve already reached the peak of their potential.
Meta AI Chief Yann LeCun has said that he’s no longer interested in LLMs, and is looking to next-generation AI architectures that’ll be able to better model the real world. He says that these new AI architectures should enable AI to think more like humans, with persistent memory and the ability to think through complex problems.
“I am not interested anymore in LLMs,” LeCun said at the NVIDIA GTC 2025 event.”They are just token generators and those are limited because tokens are in discrete space. I am more interested in next-gen model architectures, that should be able to do 4 things: understand physical world, have persistent memory and ultimately be more capable to plan and reason,” he added.
This isn’t the first time that LeCun has expressed the limitations of LLMs. LLMs largely learn through text data, and he’s said that text data alone is insufficient for achieving human-level AI. “A typical large language model is trained with something on the order of 20 trillion tokens or words,” he’d recently said. “That’s about 10 to the power 14 bytes; one with 14 zeros behind it. It’s an enormous amount of information,” he’d said.
“But then you compare this with the amount of information that gets to our brains through the visual system in the first four years of life, and it’s about the same amount,” he continues. “In four years, a young child has been awake a total of about 16,000 hours. The amount of information getting to the brain through the optic nerve is about 2 megabytes per second. Do the calculation and that’s about 10 to the power 14 bytes. It’s about the same. In four years a young child has seen as much information or data as the biggest LLMs,” he’d said.
“(This means that) we’re never going to get to human-level AI by just training on text. We’re going to have to get systems to understand the real world,” he’d said.
LeCun has now said that he’s not interested in LLMs any more. Indeed, Meta itself hasn’t released a new version of Llama in a while, and DeepSeek and other companies have taken over the mantle of releasing the top open-source models. LeCun’s approach seems to be in contrast to OpenAI’s, which has consistently maintained there’s no wall in scaling LLM capabilities, and has instead been coming up with add-ons to LLM training, like test-time compute, to increase their capabilities. It remains to be seen which approach works out, but it appears that Meta is bearish on LLMs, and seems to believe that it’ll take a new technological breakthrough to dramatically improve the performance of current AI systems.