Reports Indicate TSMC to Tighten Scrutiny on Chinese AI Chip Clients; Potential Revenue Impact Between 5% to 8%
November 11, 2024 -- Reports suggest that TSMC has notified its Chinese clients of a temporary suspension on shipments of advanced AI chips produced using 7nm and below process nodes. Additionally, Chinese clients seeking future advanced process projects will be subject to strict review to ensure the chips are not used for AI or other restricted purposes. TrendForce posits that if the policy takes effect, it could impact TSMC’s revenue performance and utilization rates of its 7nm and below advanced process capacity, while also affecting the future development of China’s AI industry.
The scrutiny follows the revelation that TSMC’s 7nm technology was used in Huawei’s 910B chip. This raised concerns over potential violations of US export regulations, along with allegations that certain companies may have acted as intermediaries to place advanced AI chip orders on behalf of Chinese firms. In response, TSMC is re-evaluating its KYC process and plants to improve its client approval standards, expand product review criteria, and potentially impose additional restrictions on chip size and HBM usage.
TrendForce notes that TSMC is highly likely to implement these measures in the near term, although the full impact will depend on whether the U.S. Department of Commerce issues additional export control regulations or adds specific customers to the Entity List.
For the first three quarters of 2024, TSMC’s revenue distribution shows that advanced nodes accounted for 67% of its total revenue—a key revenue source. However, TrendForce highlights that the major clients for TSMC’s 7/6n, 5/4nm, and 3nm processes are primarily from the U.S., Europe, and Taiwan. Consequently, even if regulatory actions impact business some Chinese clients may be lost, TrendForce expects other customers to offset this loss, limiting the potential effect on advanced process utilization rates.
TSMC’s revenue from China has remained steady at 11% to 13% for the full year of 2023 and the first three quarters of 2024. If regulatory scrutiny of TSMC’s advanced processes intensifies, or if certain Chinese clients are added to the Entity List—particularly affecting Chinese AI-related IC design companies, IP providers, third-party design services, or other businesses that depend on TSMC’s advanced processes for project initiation, tape-outs, and mass production—TSMC could face a revenue impact of approximately 5% to 8%. However, strong global demand for AI chips and TSMC’s planned price increases for advanced process clients are expected to help mitigate some of this impact.

For more information on reports and market data from TrendForce’s Department of Semiconductor Research, please click here, or email the Sales Department at SR_MI@trendforce.com
For additional insights from TrendForce analysts on the latest tech industry news, trends, and forecasts, please visit https://www.trendforce.com/news/
Related Semiconductor IP
- LLM AI IP Core
- RISC-V-Based, Open Source AI Accelerator for the Edge
- Dataflow AI Processor IP
- AI DSA Processor - 9-Stage Pipeline, Dual-issue
- RISC-V AI Acceleration Platform - Scalable, standards-aligned soft chiplet IP
Related News
- Sofics Tapes Out Test Chip on TSMC 4nm Process with Novel ESD IP and Low-Power 1.8V and 3.3V GPIO Solutions
- Cadence and Google Collaborate to Scale AI-Driven Chip Design with ChipStack AI Super Agent on Google Cloud
- Analog Bits to Demonstrate IP Portfolio on TSMC 3nm and 2nm Processes at TSMC 2025 Technology Symposium
- Synopsys and TSMC Usher In Angstrom-Scale Designs with Certified EDA Flows on Advanced TSMC A16 and N2P Processes
Latest News
- Imec unlocks fourfold UWB range extension using world-first narrowband receiver chip compliant with IEEE 802.15.4ab standard
- Alliance for Open Media Releases AV2 Codec, Advancing Next-Generation Open Video Coding
- VeriSilicon Drives Commercial Adoption of AV2 Across Next-Generation Video and Streaming Applications
- Cadence Announces Collaboration with Intel Foundry to Accelerate Intel 14A Process Optimization for HPC and Mobile Designs
- Menta and Presto Engineering Announce Strategic Collaboration to Accelerate Adaptive ASIC Architectures with Embedded FPGA Technology