TIME's 100 Most Influential in AI: What the Shift from Parameter Scaling to Real-World Deployment Signals
The latest TIME 100 AI list highlights a shift from pure algorithmic competition to open-source ecosystems and physical deployment, marking the commercialization phase for humanoid robots.
According to recent reports, the latest TIME 100 Most Influential People in AI list has been released. The United States still accounts for about 70% of the list, while nine individuals from China were selected. Alongside familiar names like Sam Altman, the inclusion of Eddie Wu (CEO of Alibaba Group and Alibaba Cloud) and the leadership behind Agibot (a prominent Chinese humanoid robotics startup) is particularly noteworthy. This is not just an honor roll; it releases a strong signal: the evaluation criteria for AI are undergoing a fundamental shift.
Shift in Evaluation Criteria: From 'Biggest Brain' to 'Physical Extension'
Over the past few years, the core narrative in the AI circle has been 'whose model has the most parameters' and 'who scores highest on benchmarks.' But the logic behind this year's list has changed. Eddie Wu was selected for promoting AI open-source initiatives, while the leadership behind Agibot represents the rise of embodied intelligence.
The global focus of AI is shifting from 'hyper-competition in pure algorithms' toward 'open-source ecosystems and real-world physical deployment.' Rather than keeping technology locked away, opening the foundational models to various industries is proving more valuable. Furthermore, giving AI a physical body to perceive reality and take action is seen as the ultimate form. For the general public, AI is no longer just a chatbot on a phone; it is transforming into a physical assistant that can help pick up packages or tighten screws on a factory assembly line.
Geeks vs. Managers: The Commercial Reality of the Robotics Track
Zhihui Jun (Peng Zhihui), a highly popular Chinese tech influencer and robotics engineer, has long been a top figure in the tech community. However, TIME accurately captured the masterminds behind his commercial ventures. Here is a very realistic case comparison: in the lab, a tech geek can make a robot perform a perfect backflip or precisely pick up a raw egg. But when it is sent to an automotive assembly line, it needs to continuously tighten 100,000 screws without error, and the total manufacturing cost must be kept within the budget of an average family buying a commuter car.
We can imagine a specific scenario: a tech geek excitedly demonstrates a new algorithm in the lab, but the procurement director pours cold water on the idea while holding the BOM (Bill of Materials): 'The cost of this joint motor must be cut by 30%, otherwise the retail price won't drop below $15,000, and it simply won't sell.'
Relying solely on geeks staying up late to tweak parameters is not enough. The humanoid robotics track is officially transitioning from a 'tech geek show' to a 'commercial operations' phase. It requires 'managers' who understand supply chains and cost control to obsess over the procurement price of every single part and truly get the production line running.
Three-Step Asymmetric Competition: The Underlying Logic of the US and China AI Ecosystems
Zooming out, this list reflects the differences in the AI paths of the US and China. The US accounts for 70%, with strengths in foundational algorithms and computing chips; the nine Chinese selections highlight their position in open-source ecosystems and hardware manufacturing. This is essentially a case comparison of two ecosystems: the US is building the 'highways' (foundational large models), while China is building the 'cars running on them' (hardware) and establishing 'free and open traffic rules' (open-source).
To understand this asymmetric competition, it can be broken down into three steps: Step one is the asymmetry in infrastructure and manufacturing, where the US focuses on the foundational layer while China leverages its hardware supply chain advantages; step two is the flywheel effect of the open-source ecosystem, attracting global developers to refine applications through openness; step three is the occupation of physical terminal scenarios, using massive real-world data to feed back into and improve algorithms.
If AI terminals in the physical world experience a major boom in the coming years, China's robust hardware manufacturing and open-source communities could become the biggest amplifiers for technology deployment. Of course, shortcomings in foundational computing power still need time to be addressed, and this long race remains to be seen.
Crossing the 'Valley of Death' in Mass Production: Supply Chain Risks Behind the Boom
Despite the favorable trend toward real-world deployment, we cannot ignore the hidden risks. The biggest risk for humanoid robots currently lies in the 'Valley of Death' of mass production and the loss of cost control in the supply chain. Going from one unit in the lab to 10,000 in the factory is not just an addition of numbers; it is a hellish test of yield rates, quality control, and supply chain resilience. If a core component encounters a production bottleneck or price fluctuation, the entire commercialization process could be derailed.
It is worth noting that capital markets often overvalue 'technological breakthroughs' while lacking patience for the 'hard work of manufacturing.' If companies only tell stories without balancing the BOM, they can easily face a broken capital chain on the eve of mass production.
Key Takeaway: AI's testing ground has changed. It is no longer about 'who has the smartest brain,' but rather 'who has the most agile hands and feet, who offers the most open blueprints, and who calculates the balance sheet most precisely.'