An industrial robot arm that picks up small parts moves, and behind it, artificial intelligence (AI) server racks are on display. Digital screens analyzing factory operations and logistics flows in real time were installed throughout the exhibition hall.
On the 4th at Nangang Exhibition Center in Taipei, Taiwan. At Computex 2026, where Nvidia graphics processing units (GPU) and AI servers filled the floor, competition over the "AI Factory" that uses AI for the consolidation of factories, warehouses, and industrial robots was in full swing. This year's Computex has evolved beyond simply pitting chip performance against one another to a stage showing how AI is changing actual manufacturing sites.
This trend was evident from the opening ceremony. Taiwan President Lai Ching-te, in the Computex 2026 opening address on the 2nd, announced the "10 major new AI infrastructure projects," which include training 500,000 AI experts by 2040 and expanding the supply of infrastructure such as water, power, and land. The plan is to strengthen, at a national level, the foundation of the AI industry beyond semiconductor manufacturing in response to growing AI demand and develop Taiwan into a key global AI industry hub.
Taiwanese corporations are accelerating the implementation of AI factories by touting an "AI manufacturing ecosystem" consolidated from design to production and infrastructure. Centered on TSMC, Taiwan has clusters of semiconductor production, AI servers, and industrial automation companies, earning it a reputation for having a favorable environment to build an AI manufacturing ecosystem.
In the exhibition hall, next-generation manufacturing infrastructure using AI was becoming reality. The booth of Foxconn, the world's largest electronics contract manufacturer, was a prime example. The monitors at the front of the booth displayed a Digital Twin that recreated the inside of a factory in virtual space. AI analyzed production equipment, logistics flows, and the status of industrial robot operations in real time. Whereas past smart factories were at the level of consolidation of automated equipment, they have now evolved into a stage where AI analyzes and optimizes production planning and operational efficiency.
Power and industrial automation corporation Delta also put its AI-based smart factory front and center. The booth demonstrated a solution in which AI analyzes the condition of factory equipment in real time and predicts anomalies. The key is to automatically optimize even energy use to raise productivity and power efficiency at the same time.
Advantech and GigaByte showcased foundational technologies that make AI work on actual manufacturing floors. This "Edge AI" platform analyzes on site, without sending the data collected by in-factory cameras and sensors to an external data center, and controls production equipment and robots. They said it boosts data processing speed and reduces latency, making it suitable for manufacturing environments where real-time control is crucial.
This also aligns with "physical AI," a key theme of this year's Computex. AI is evolving beyond merely generating information on screens to directly moving robots, production equipment, and logistics systems.
Jensen Huang, Nvidia chief executive officer (CEO), repeatedly mentioned AI factories during Computex. Huang said that not only data centers that train AI but also the importance of manufacturing infrastructure where AI is used on actual industrial sites is growing. Nvidia is working with Taiwanese manufacturing corporations such as Foxconn, Advantech, and Delta to build an ecosystem that enables the consolidation of AI servers, industrial computers, robots, and factory automation systems.
An industry official said, "The center of gravity in AI competition is shifting from big tech's data center buildouts to process innovation on manufacturing floors," adding, "It's a stage to prove how AI controls the hardware of actual industrial sites and changes productivity."