Qiming Venture Capital Releases “2026 AI Top 10 Prospects,” Covering Foundational Models, Embodied Intelligence, AI Infrastructure, and AI Applications

Deep Tide TechFlow message: On July 19, at WAIC’s “Qiming Venture Capital · Entrepreneurship and Investment Forum,” Qiming Venture Capital released the “2026 Qiming Venture Capital Top 10 AI Outlooks,” covering four areas: foundational models, embodied intelligence, AI infrastructure, and AI applications.

Among them, in the area of foundational models, Qiming Venture Capital expects that, over the next 12 to 24 months, leading models will internalize most of the capabilities of “plug-ins,” and multimodal models will further evolve toward modeling interactive worlds.

In the area of embodied intelligence, the effective data scale of leading robotics companies is expected to rise from the “ten-thousand-hours level” in 2025 to the “million-hours level” in 2027. Dexterous hands and tactile sensing technologies will mature faster, and costs will continue to decrease.

In the area of AI infrastructure, it pointed out that the focus of AI computing demand is shifting toward inference. Over the next two years, AI infrastructure may continue to face structural shortages, and secure and trustworthy AI will also upgrade from an optional option to a required one.

In the area of AI applications, over the next 12 to 24 months, AI application business models will accelerate toward result- and value-based pricing. Commercialization will focus more on vertical scenarios and high-paying users, and AI-Native organizations will move from concept to proof.