In a major leap for global technology, Taiwan Semiconductor Manufacturing Company (TSMC), the world's leading chipmaker, has announced plans to begin mass production of its next-generation "2-nanometre" (2nm) processors by the end of this year. This advancement promises to redefine computing power, energy efficiency, and the capabilities of artificial intelligence systems worldwide.
The Power of 2 Nanometres: What's the Big Deal?
For decades, the relentless drive in semiconductor technology has been to pack more microscopic components, called transistors, onto a single silicon chip. This miniaturization is the engine behind every technological jump, from the evolution of smartphones to the rise of generative AI tools like ChatGPT. The new 2nm technology represents the next frontier in this journey.
According to computing giant IBM, these advanced chips will lead to faster laptops, a reduced carbon footprint for massive data centres, and allow self-driving cars to process information and identify objects more quickly. For the booming field of artificial intelligence, the impact will be twofold. "This benefits both consumer devices -- enabling faster, more capable on-device AI -- and data centre AI chips, which can run large models more efficiently," explained Jan Frederik Slijkerman, a senior sector strategist at Dutch bank ING.
The Elite Club of Chipmakers and Geopolitical Tensions
Crafting chips at the 2nm scale is a monumental challenge. It demands cutting-edge lithography machines, profound expertise in the production process, and investments running into tens of billions of dollars. Consequently, only a handful of companies are in this elite race. TSMC currently leads the pack, with South Korea's Samsung and US-based Intel working to improve their production yields and attract large-scale customers. Japanese chipmaker Rapidus is also constructing a plant in northern Japan, targeting 2nm mass production by 2027.
TSMC's path, however, has not been without hurdles. In a stark reminder of the high stakes, Taiwanese prosecutors in August charged three individuals with stealing trade secrets related to the 2nm process to benefit a Japanese equipment maker. The case was described as involving "critical national core technologies vital to Taiwan's industrial lifeline."
Geopolitics and US-China trade tensions also cast a shadow. Reports suggest TSMC, whose client list includes tech giants like Nvidia and Apple, will exclude Chinese-made equipment from its 2nm production lines to avoid potential US sanctions. Furthermore, the company has stated plans to accelerate the production of advanced chips, including 2nm technology, at its US facilities, with a current target set for "the end of the decade."
How Small is 2nm and What Comes Next?
The term "2-nanometre" is more a marketing benchmark for transistor density than a literal measurement of size. For perspective, a single atom is about 0.1 nanometres across. IBM states that 2nm designs can cram up to 50 billion transistors—each smaller than a virus—onto a chip the size of a fingernail. A higher density means either a more powerful chip of the same size or a smaller chip with equivalent performance, leading to greater energy efficiency.
The race does not stop here. TSMC is already developing even more advanced "1.4-nanometre" technology, with mass production rumoured for around 2028. Samsung and Intel are hot on its heels. TSMC began high-volume 3nm production in 2023 and is reportedly building a new factory in Taichung, Taiwan, dedicated to the future 1.4nm chips.
Despite the immense technical and financial hurdles, the industry sees 2nm chips as foundational for the future. As Rapidus highlighted, they are "ideal for AI servers" and will "become the cornerstone of the next-generation digital infrastructure." The mass production of these chips in 2025 marks the beginning of a new, more powerful, and efficient era for the digital world.
