rotative
arstechnica ai · 2026年7月13日

為什麼World Labs認為空間智能將把AI從LLM泡沫中拯救出來

像world labs這樣的世界模型正受到關注,因為fei-fei li和yann lecun等專家預測,ai將從llm轉向能夠模擬物理現實的系統。

read in English · 简体中文 · 日本語 · 한국어

as the ai landscape pivots from pure language processing, world models—systems that simulate physical reality—are drawing massive funding and research attention. led by figures like fei-fei li of world labs and yann lecun, this sector aims to ground ai in spatial intelligence for robotics and asset generation. this movement challenges the current dominance of llms, with some lab leaders predicting the llm bubble will burst. commercial applications are already emerging from runway and google deepmind, signaling a tangible industry shift toward physically grounded ai.

read at source ↗

← rotative