The Power of TSMC’s A14 Chip Process for Future AI

As AI workloads continue to drive unprecedented demand for advanced semiconductors, the global chip manufacturing industry faces mounting pressure to deliver more powerful and energy-efficient solutions.

In this high-stakes environment, Taiwan Semiconductor Manufacturing Company (TSMC) maintains its position as the critical linchpin in the global technology supply chain, producing chips for companies including Apple, AMD, Nvidia and Qualcomm.

The semiconductor manufacturing market has become increasingly complex since 2022, with geopolitical tensions and supply chain vulnerabilities prompting major economies to invest billions in domestic chip production capabilities.

Now, TSMC has unveiled its next-generation A14 process technology at its North America Technology Symposium in Santa Clara, California.

This is a step forward from the company's N2 process, scheduled to enter production later this year – as the A14 technology is designed to accelerate AI capabilities through improved computing performance and power efficiency – addressing growing concerns about the energy consumption of large AI systems.

What is TSMC’s A14 process?

The A14 process, scheduled to enter production in 2028, is currently under development with yield performance ahead of projected timelines.

TSMC reports that compared to the N2 process, which will begin volume production later this year, A14 will deliver up to 15% speed improvement at equivalent power consumption, or up to 30% power reduction at the same speed, alongside more than 20% increase in logic density.

TSMC Chairman and Chief Executive Officer Dr C.C. Wei says: “TSMC's cutting-edge logic technologies like A14 are part of a comprehensive suite of solutions that connect the physical and digital worlds to unleash our customers' innovation for advancing the AI future.”

The impact of TSMC’s expanded Chip on Wafer technology 

TSMC also continues to advance its Chip on Wafer on Substrate (CoWoS) technology to address increasing demands for logic processing and high-bandwidth memory in AI applications.

The company plans to commence volume production of 9.5 reticle size CoWoS in 2027, enabling integration of 12 or more high-bandwidth memory stacks in a package together with TSMC's logic technology.

Following its System-on-Wafer (TSMC-SoW) technology introduction in 2024, TSMC has now announced SoW-X, a CoWoS-based offering to create wafer-sized systems with computing power 40 times greater than current CoWoS solutions.

Volume production for this technology is planned for 2027.

The company has additionally developed complementary solutions to enhance the computing efficiency of its logic technologies.

These include silicon photonics integration with the company's Compact Universal Photonic Engine, N12 and N3 logic base die for HBM4 (fourth-generation high-bandwidth memory) – and a new Integrated Voltage Regulator for AI applications with five times vertical power density delivery compared with separate power management chips on circuit boards.

N4C RF technology targeting multiple sectors for enhanced AI capabilities

TSMC is additionally supporting AI on edge devices and its need for high-speed, low-latency wireless connectivity to move massive data with N4C RF, the latest generation of TSMC’s radio frequency technology.

Smartphones
For smartphone manufacturers seeking to incorporate AI functions into edge devices, TSMC is introducing N4C RF.

This offers 30% power and area reduction compared with N6RF+, making it suitable for incorporating more digital content into radio frequency system-on-chip designs to meet requirements of emerging standards such as WiFi8 and AI-enhanced True Wireless Stereo. Risk production is scheduled to begin in the first quarter of 2026.

The automotive industry
The chipmaker is also addressing requirements for automotive applications, particularly Advanced Driver Assistance Systems and Autonomous Vehicles, which require substantial computing power without compromising on automotive-grade quality and reliability.

TSMC's N3A process is in the final stage of AEC-Q100 Grade-1 qualification, with continuous defect improvement to meet automotive defective parts per million requirements. N3A is entering production for automotive applications as part of a technology suite for future software-defined vehicles.

IoT applications
For Internet of Things (IoT) applications, which increasingly incorporate AI functionality while maintaining strict power efficiency requirements, TSMC has commenced production of its previously announced ultra-low power N6e process.

The company is now targeting N4e to further improve power efficiency for future edge AI applications.

“Our customers constantly look to the future, and TSMC's technology leadership and manufacturing excellence provides them with a dependable roadmap for their innovations.” says Dr C.C. Wei.

Explore the latest edition of AI Magazine and be part of the conversation at our global conference series, Tech & AI LIVE. 

Discover all our upcoming events and secure your tickets today.

Also sign up to our free weekly newsletter for the latest insights and stories straight into your inbox.

AI Magazine is a BizClik brand