[World Model Funding Cools Down: Only 5 Rounds in October, Investors Become More Selective] Beating AI News Flash: World models, which were highly sought after by investors in the first half of the year, have cooled down in the second half. A recent interview with several leading investors and startup fundraising executives by QuantumBit revealed that an increasing number of world model projects are struggling to even obtain term sheets (TS). Several investors shared a highly similar list of teams they were willing to invest in, with only about four or five teams listed. Incomplete statistics from Cailian Press Venture Capital show that over 100 funding rounds related to physics AI and world models were planned for 2026, but only 5 were held in October. Wang Sheng, a partner at InnoFund, revealed that top VCs began slowing their investment pace in May and June. The number of project deals closed in July and August declined consecutively. Since funding announcements are usually delayed by several months, many of the seemingly bustling funding news in the second half of the year are actually continuations of deals made in the first half. World models themselves also face challenges in technology implementation. There's no universally accepted definition for this concept; generating realistic videos, simulating physical laws, and predicting robot movements are all referred to as world models. World Labs, founded by Fei-Fei Li, once categorized them into renderers, simulators, and planners to differentiate between different technological approaches. In May of this year, a study tested nine video generation models using 319 sets of physical conditions, with the highest overall score for a single pair being only 51.7%. Even when the generated footage appeared realistic, it often failed to accurately reflect the effects of changes in conditions such as friction and object mass. However, capital hasn't completely withdrawn. Interviews conducted by QuantumBit revealed that a few projects with top-tier research teams, technological achievements, or resources from large companies are still able to secure substantial funding. Compared to the hype surrounding concepts in the first half of the year, investors now place greater emphasis on whether the team can produce tangible results. [Original Link]