Svmuu News: Serenity, known as the “White-Haired Stock Guru,” posted on X, stating that, based on capital flows in China’s private equity and venture capital (VC) markets, institutions are currently pouring massive amounts of capital into areas related to Physical AI and World Models.
Data shows that capital allocation is roughly as follows: large language models (LLMs) at approximately $23.56 billion, AI infrastructure and technology at approximately $15.74 billion, embodied intelligence/ Physical AI at approximately $13.36 billion, AIGC applications at approximately $8.79 billion, while autonomous driving and the other top 20 sub-sectors total approximately $3.82 billion (the metrics are not directly comparable).
Serenity points out that early-stage funding for pure foundational models has largely dried up, with capital now concentrated on established industry leaders and the “world model” direction. He expects this trend to emerge in the United States as well, with capital potentially concentrating further on leading companies such as Anthropic and OpenAI in the future. Regarding AIGC applications, Serenity believes that commercialization in this sector is relatively mature, but no clear winner has emerged yet, with a fragmented competitive landscape in both the U.S. and Chinese markets.
Overall, Serenity concluded that AI investment continues to flow into infrastructure and the semiconductor supply chain, while capital is rapidly shifting toward physical AI and embodied intelligence; however, the world model sector still lacks direct investment targets.