From Innovation Cycle to Industrial-Scale Cycle
台積電 Vice President of Business Development 袁立本 attended the Taiwan Semiconductor Industry Association annual meeting on the 29th and delivered a speech on Taiwan's competitive advantages and challenges in light of AI development trends. He noted that the launch of ChatGPT in Q4 2022 marked the first widespread use of large language models by humans, and that before 2025 the AI industry was in an innovation cycle focused on large language model breakthroughs, algorithm exploration and early agentic AI proof-of-concept, a stage of traditional and AI hybrid operation. Starting this year, AI has entered an industrial-scale cycle, no longer a laboratory stage, but entering large-scale deployment, including infrastructure build-out and large-scale applications such as enterprise-grade AI agents, dedicated data center construction and embodied AI robots.
He said capital expenditure and industry adoption have seen explosive growth, deeply felt across Taiwan's industry chain, and the speed of the explosion has also brought unprecedented challenges. The supply chain must have very high execution capability to simultaneously complete system validation and rapid deployment. To support a hundred-fold surge in computing power, substrates, printed circuit boards, power supplies and cabinet-related high-speed optical interconnects should be deeply integrated; if any one bottleneck occurs, the cabinet will get stuck, and the industry should adopt full-stack co-innovation from chips to cabinets.
Taiwan's Moat and Challenges
袁立本 stressed that Taiwan has the most complete and dense industry cluster, not only the leading wafer foundry, but also a global AI system integration solution provider. From wafer manufacturing, packaging, memory, substrates and thermal interface materials to factory shipment, it has one-stop manufacturing capability with 大量 production validation, difficult for any other region in the world to replicate in the short term, which is Taiwan's moat. The challenge for Taiwan's industry is whether it can provide real-time solutions for system and performance bottlenecks, how to capture higher service value from AI data centers and applications beyond hardware manufacturing, and how to scale AI and physical AI applications.
Others include how to cultivate domestic talent and attract overseas talent, how to supply power resources needed for AI data center construction and chip capacity expansion, how to support rapid capacity expansion in land acquisition and regulatory adaptation, and how to respond to the impact of geopolitics on overseas deployment and supply chain regionalization.
Helping Innovation Leap to Neural Hub
To this end, 袁立本 proposed recommendations to assist創新建議, shifting from a passive foundry model to early-stage technology co-design, and promoting joint experimental platforms to break through technological bottlenecks together, while promoting open-source tools, setting standard formats and governance norms, and building a secure cross-industry data platform. He also suggested upgrading talent cultivation, expanding AI professional training, building a national-level AI talent pool, improving regulations and standards, accelerating AI product certification and intellectual property protection mechanisms, providing funding and tax incentives, and deepening local self-development and resilience of the supply chain, with the aim of propelling Taiwan to become a global AI neural hub.