TSMC has a high degree of dominance in advanced logic chip manufacturing. (Illustration, AFP)
[Financial Channel/Comprehensive Report] Competition in the AI chip market continues to heat up. Huida currently still holds the leading position in data center computing chips. However, as major AI ultra-large-scale cloud service providers gradually join the fray, there are still variables in the market who will win in the future. However, no matter which chip company wins in the end, there is one company that may continue to benefit, and that is TSMC. Foreign media "The Motley Fool" analyzed that TSMC has established an almost unshakable competitive advantage in the field of AI data centers by virtue of its huge production capacity, advanced process technology and long-term cooperative relationships with customers.
Reports indicate that the number of data center computing units currently produced by Huida is still far ahead of its competitors, and it has established a first-mover advantage by being the first to enter the AI computing market. However, AMD is also actively exploring the GPU market, and Broadcom is cooperating with AI ultra-large-scale cloud service providers to develop customized AI chips designed for specific workloads, which may put pressure on the GPU market share of Huida and AMD.
Please read on...
However, this AI chip competition has one thing in common: TSMC is behind the major chip companies. In addition to producing chips for Huida, TSMC also provides logic chip manufacturing services to AMD, Broadcom and other companies, making it an important common supplier in the AI computing industry chain. According to TrendForce estimates, TSMC accounted for 72.5% of global foundry industry semiconductor revenue in the second quarter of this year, far ahead of other competitors; Intel did not even make it into the top 10.
Market participants believe that TSMC’s real advantage is not just its market share, but its advanced process capacity and customer relationships accumulated over many years. Even if AI chip designers want to change suppliers, they must consider factors such as process maturity, production capacity scale, yield rate, and long-term cooperation. Therefore, it is very difficult for competitors to seize a large amount of TSMC's market share at once. Even if some orders flow to other foundries, most of the manufacturing needs for AI computing chips are likely to remain in the hands of TSMC.
More importantly, capital expenditures for AI data centers are still expanding rapidly. In the long term, Huida estimates that global data center annual capital expenditures may reach US$3 trillion to US$4 trillion by 2030. This means that as long as AI infrastructure continues to expand, demand for GPUs, customized AI chips and related computing equipment may increase simultaneously, and TSMC, as a major chip manufacturer, will also benefit from this.
TSMC itself is expanding production capacity for the next wave of demand. In the second quarter of this year, the company announced an additional US$100 billion investment in production facilities in Arizona, showing that TSMC still has high expectations for future demand for AI and advanced chips. Huge capital expenditure also creates another competitive threshold, because foundry requires a large amount of long-term investment in the construction of advanced processes and production facilities, and it is not possible to catch up in the short term.
Therefore, while Huida, AMD, Broadcom and other AI chip companies compete for the market, TSMC may become a supplier that benefits the entire AI chip industry. The market is still unsure which AI chip design company will gain the ultimate advantage in the future, but as long as the AI data center continues to expand, the demand for chip manufacturing will be difficult to disappear. For investors, this also brings more attention to TSMC’s role in the AI industry chain.
Grasp the economic pulse with one hand. Click here to subscribe to Free Finance Youtube channel
