Yann LeCun: LLMs Are Nearing the End, but Better AI Is Coming (2025)
Yann LeCun argues that while large language models (LLMs) have made impressive strides, they are fundamentally limited by their inability to model the continuous, high-dimensional nature of the real world and will soon reach their developmental limits. He contends that true AI advancement requires systems capable of reasoning, planning, and world modeling akin to human 'System 2' cognition. Such future AI systems must operate with abstract representations, goal-driven learning, and sensorimotor experience to achieve robust, general intelligence. Despite current commercial reliance on LLMs, LeCun believes the next breakthrough will come from architectures that simulate the physical world more effectively.
Opening excerpt (first ~120 words) tap to expand
Yann LeCun always reminds me of the very best of Bell Labs' scientists and engineers—a unique breed of individual, fiercely independent of thought and action, who thrive within company structures that typically value obedience and conformance to the corporate mantra and goals. In my experience, corporate parents only tolerate such independence of thought and deed as the price to pay for attracting the best and brightest minds that are the fundamental catalyst for disruptive innovation. But I find that this understanding and "détente" is increasingly rare in modern business culture in which uniformity and alignment seem to be prized above all, possibly due to the hyperbolic nature of our times and the denigration of dissenting voices.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Newsweek.
Discussion
0 commentsMore from Newsweek
