TSMC used its Q1 2026 earnings call to do two things at once: celebrate a strong quarter and underline why it believes its technology roadmap keeps it ahead in the fast-moving foundry race.
The company reported Q1 FY26 revenue of $35.9 billion, a 6.4% increase quarter over quarter. Chairman and CEO C.C. Wei used the call to discuss TSMC’s progress on advanced process nodes, its outlook on supply chain stability, and how it views rising competition in contract chip manufacturing.
One of the most talked-about topics was Intel’s growing push into foundry services, including its Terafab plans with Tesla. Wei described Intel as a formidable competitor and made it clear TSMC isn’t dismissing any rival efforts. At the same time, he emphasized a key reality of semiconductor manufacturing: even for major companies, success in foundry work takes years of disciplined execution.
Wei pointed out that building a new chip fab typically takes 2 to 3 years, and bringing it to meaningful production levels takes another 1 to 2 years after that. In other words, there are no shortcuts. That timeline matters for projects like Tesla’s Terafab, which is expected to start with trial output of around 3,000 wafers per month. Volume production takes time, so companies expanding manufacturing capacity often still rely on established semiconductor partners during the ramp.
Intel, meanwhile, is advancing its own node roadmap with products such as 18A and 14A. The 14A node in particular is being positioned as a major step for Intel’s foundry ambitions, with expectations that it could attract significant customers over time.
The conversation also touched on advanced packaging, a crucial battleground as AI accelerators and high-performance computing chips grow more complex. When asked about Intel’s EMIB packaging approach, Wei said TSMC is already supplying what it describes as the industry’s largest reticle-sized packages used in leading chips. Still, he called EMIB an attractive technology and framed competition in packaging as a positive, because more options help customers make better platform decisions.
On the manufacturing side, TSMC highlighted that its N2 process has been in high-volume manufacturing since Q4 2025 and is delivering good yields. N2 introduces the company’s first-generation nanosheet transistor technology and is ramping in phases across TSMC’s Hsinchu and Kaohsiung sites. Demand is being driven by both smartphone chips and high-performance computing, especially AI-focused workloads.
TSMC also expects the N2 family to have a long lifespan, supported by ongoing enhancements such as N2P and A16. The company believes N2 will be an important platform for the current wave of AI growth, including the continued momentum behind agentic AI, while also remaining central for consumer products. Future PC and server platforms from major designers are expected to use variants like N2P, reinforcing how critical this node is likely to be across multiple categories.
Wei also addressed supply chain resilience, noting that TSMC is preparing for potential disruptions involving key materials, including vital gases and chemicals such as LPG. With geopolitical tensions affecting global supply routes, the company said it currently has about three months of LPG inventory available.
Looking further ahead, TSMC shared fresh details about its next-generation A14 (1.4nm-class) process, described as a “full-node stride” beyond N2. A14 will use a second-generation nanosheet transistor structure and targets the core needs of modern computing: higher performance and better energy efficiency.
Compared with N2, TSMC projects A14 can deliver a 10% to 15% speed increase at the same power, or a 25% to 30% power reduction at the same speed, along with roughly a 20% improvement in chip density. Development is said to be on track, with strong customer interest from both smartphone and HPC segments. Volume production is currently scheduled for 2028.
Overall, the message from TSMC’s earnings call is straightforward: the company sees competition ramping up across nodes, packaging, and capacity expansion, but it believes its manufacturing scale, process roadmap, and execution discipline keep it well-positioned for the next phase of AI-driven and high-performance computing growth.
