The "AI Godmother" Fei-Fei Li's spatial intelligence's first model is born! An image generates an interactive 3D world

According to media reports, on December 3 local time, the spatial intelligence startup World Labs, co-founded by "AI Godmother" Fei-Fei Li, showcased its first achievement—an AI system that can generate a 3D world from a single image and a sentence, dubbed the "virtual world generator." World Labs refers to this as the first step towards spatial intelligence.

Zhongtai Securities stated that a new generation of artificial intelligence technology and industrial transformation, represented by large models and generative methods, is advancing rapidly. From the text-to-text capabilities of Chat GPT, to the text-to-image capabilities of DALL-E, and then to the text-to-video capabilities of Sora, the "aesthetic of violence" continues to push the ceiling of technology. Multimodality has become a consensus development trend; after text, code, images, and videos, Zhongtai Securities believes the next modality likely to achieve breakthroughs is 3D. Currently, overseas exploration of AI + 3D technology is mainly divided into industrial scene exploration and non-industrial scene exploration. From an industrial perspective, it is suggested to continuously track and pay attention to the progress in the field of text-to-3D modeling.