Yann LeCun has publicly criticized Geoffrey Hinton's recent views on Large Language Models (LLMs), suggesting Hinton's embrace of LLMs as nearing human-level intelligence is a desire to retire and focus on AI's dangers. LeCun argues that LLMs are useful tools but not a path to Artificial General Intelligence (AGI), contrasting them with his own work on world models and Joint Embedding Predictive Architectures (JEPA). He also revealed his departure from Meta's FAIR to found AMI (Advanced Machine Intelligence) to focus on AI for the real world, as Meta's strategic shift towards LLM competition sidelined his research. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Yann LeCun's critique of LLMs and his new venture focusing on world models may influence research directions and investment in AI beyond current LLM trends.
RANK_REASON This cluster features an opinion piece by a prominent AI researcher criticizing another prominent researcher's views on LLMs and discussing his own research direction and new venture.