Qiming Venture Capital hosted the "Qiming Venture Capital · Entrepreneurship and Investment Forum – The Evolutionary Journey from Computing Power Origins to Application Implementation" at the 2026 World Artificial Intelligence Conference in Shanghai.
Public information indicates that Qiming Venture Capital began systematically investing in the artificial intelligence sector starting in 2013. To date, the firm has invested in over 100 AI projects with a total investment exceeding 12 billion Chinese yuan, having successfully identified and fostered the growth of several innovative benchmark companies.
During the opening keynote, Zhou Zhifeng, Managing Partner at Qiming Venture Capital, delivered a speech titled "Riding the Technology Wave: Finding the Commercial Inflection Point in the AI Era." The presentation centered on foundational models, embodied AI, AI infrastructure, and AI applications, outlining the firm's ten key AI predictions for 2026:
Initial Steps
Prediction One: Within the next 12-24 months, top-tier models will internalize most "plug-in" capabilities, including task planning, tool invocation, multi-agent collaboration, and some harness engineering abilities.
Prediction Two: Multimodal models will further evolve towards modeling interactive worlds, becoming a key technological pathway for AI to achieve environmental perception, long-term planning, and large-scale deployment in the physical world.
Prediction Three: The volume of effective data held by leading robotics companies is projected to surge from a scale of "tens of thousands of hours" in 2025 to "millions of hours" by 2027, with human first-person perspective data expected to dominate absolutely.
Prediction Four: Dexterous robotic hands will experience accelerated development. Hands with tactile perception will mature and see costs continue to decline, forming a complementary high-low mix with two-finger gripper solutions and gradually achieving scaled penetration in complex operational scenarios.
Core Infrastructure and Competitive Landscape
Prediction Five: The focus of AI computing power demand will shift towards inference. Supply constraints in storage, advanced semiconductor manufacturing processes, and advanced packaging will intensify layer by layer. AI infrastructure will face persistent structural shortages over the next two years, elevating computing power asset reserves to a core strategic priority for AI enterprises.
Prediction Six: AI infrastructure competition will enter a systems-level phase, spanning key areas such as chips, interconnects, cooling, and power supply. Within the next two years, computing power chips based on new architectures and super-node large-scale clusters are anticipated to emerge, aiming to achieve low-cost, high-efficiency token production.
Essential Factors for Deployment and Commercialization
Prediction Seven: Safety, trustworthiness, and ethics will join product efficacy and token cost as the three critical variables impacting large-scale enterprise AI adoption. Secure and trustworthy AI will transition from an optional feature to a mandatory requirement.
Prediction Eight: Over the next 12-24 months, the business models for AI applications will accelerate their departure from the internet-era freemium logic, shifting towards pricing based on results and value. The core metric for evaluating AI companies will evolve from user scale to the commercial value created per unit of intelligence cost.
Prediction Nine: In the coming 12-24 months, AI application commercialization will concentrate on vertical scenarios and high-paying users, with efficiency-focused applications leading the initial wave ahead of consumer-facing ones. As token costs continue to fall and interaction paradigms innovate, blockbuster AI consumer applications will gradually emerge.
Organizational Transformation
Prediction Ten: Within the next 12-24 months, the concept of AI-Native organizations will move from theory to practice, with a cohort of enterprises expected to achieve per-capita productivity several times greater than that of traditional organizations.
Comments