Unverified60% confidenceOpinionExact time
Yann LeCun believes that current large language models based on the autoregressive Transformer architecture cannot lead to AGI, and that truly general intelligence requires an entirely new world model architecture.
2
Sources
60%
Confidence
Long-term
Relevance
9/2/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedTuring Award laureate Yann LeCun believes the architecture of current language models is fundamentally insufficient to produce truly goal-directed behavior, and that safety concerns are being overstated.69% similarVerifiedWithout tool-calling capability, large language models can only operate within the language space and cannot truly affect external systems.66% similarVerifiedLarge language models are essentially stateless and have no native cross-session memory mechanism, which is rooted in the Transformer architecture itself.66% similarUnverifiedReal-world large language models and image generation tools are far from comprehensively surpassing human intelligence.66% similarUnverifiedFive years ago, large language models had not yet entered public awareness, and the capabilities of generative AI were far below current levels.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/842933API
curl https://kongchang.com/api/v1/knowledge/claims/842933MCP
get_claim(id=842933)