Morgan Stanley's in-depth perspective on the ASIC industry chain: Who are the big winners?

With the rapid development of generative AI applications, whether AI ASIC can become a viable alternative to NVIDIA GPUs has been a hot topic globally. On the 15th, Morgan Stanley released a research report titled "AI ASIC 2.0: Potential Winners," stating that ASICs, with their targeted optimization and cost advantages, are expected to gradually capture more market share from NVIDIA GPUs.

Morgan Stanley predicts that the AI ASIC market size will grow from USD 12 billion in 2024 to USD 30 billion in 2027, with a compound annual growth rate of 34%.

In this context, NVIDIA, leveraging its advantages in large language model training, will continue to dominate the market. Broadcom, Alchip, and Socionext are viewed positively. Cadence, Taiwan Semiconductor, and its supply chain partners (ASE, KYEC, etc.) will benefit from the rapid growth in ASIC design and manufacturing.

Morgan Stanley stated that the rise of ASICs does not mean the decline of GPUs. On the contrary, these two technologies will coexist in the long term, providing optimal solutions for different demand scenarios.

Will ASIC Become a Strong Competitor to NVIDIA?

With the rapid development of generative AI applications, global AI computing demand is experiencing explosive growth. The report predicts that under the basic scenario, the cloud AI semiconductor market size will reach USD 238 billion by 2027, and in an optimistic scenario, it could even reach USD 405 billion.

In this field, ASICs, with their targeted optimization and cost advantages, are expected to gradually capture more market share from NVIDIA GPUs.

Morgan Stanley predicts that the AI ASIC market size will grow from USD 12 billion in 2024 to USD 30 billion in 2027, with a compound annual growth rate of 34%.

Although NVIDIA's AI GPU performance is outstanding, Morgan Stanley believes that cloud service providers such as Google, Amazon, and Microsoft are still actively promoting ASIC design. The main driving forces behind this are twofold.

First, optimizing internal workloads. By developing custom chips, CSPs can more efficiently meet their internal AI inference and training needs.

Second, better cost performance. The report points out that while NVIDIA's GPUs have powerful computing performance, their hardware prices are high, especially during AI training processes. In contrast, ASICs have lower unit costs, particularly after large-scale use.

For example, Amazon's Trainium chip is about 30% to 40% cheaper than NVIDIA's H100 GPU for inference tasks. Google is also continuously optimizing its TPU series, with the latest TPU v6 improving energy efficiency by 67% compared to the previous generation.

Morgan Stanley emphasizes that although NVIDIA's GPUs remain the preferred choice for most CSPs, in the coming years, as ASIC designs mature, these cloud giants may gain greater bargaining power in procurement negotiations through self-developed ASICs.

Winners and Losers: Who Will Dominate the Future Market?

Morgan Stanley's report outlines the global ASIC supply chain and identifies six potential winners:

• AI GPU: NVIDIA will continue to dominate the market, especially in large-scale language model training, where its solutions remain the optimal choice.
• ASIC Suppliers: Broadcom, Alchip, and Socionext are seen as potential stocks in the ASIC market. Among them, Alchip is expected to significantly increase its market share by 2026 due to its deep collaboration with AWS.
• Electronic Design Automation Tools: Cadence is expected to achieve structural growth.
• Foundries: Taiwan Semiconductor and its supply chain partners (such as ASE, KYEC, etc.) will benefit from the rapid growth of ASIC design and manufacturing.
• Testing Services: Advantest is a leader in AI chip testing, and its focus on testing AI devices will bring significant growth.
• HBM: Samsung Electronics is the leader in the non-NVIDIA HBM market share and will benefit from the growth in ASIC demand.

In contrast, some traditional chip companies and foundries may face challenges. For example, AMD may lose more market share due to its failure to close the gap with NVIDIA in the AI GPU field. Foundries like UMC, lacking support for advanced process nodes, may also struggle to gain a foothold in the high-end AI chip market.

TCO Analysis: Is ASIC Really Cost-Effective?

Morgan Stanley compared the cost-effectiveness of ASICs and GPUs in AI training and inference tasks using a TCO model. The results show that although NVIDIA's GPUs have a clear performance advantage, ASICs have lower initial costs, making them particularly suitable for budget-constrained cloud service providers.

For instance, within the same budget, AWS's Trainium 2 can complete inference tasks faster than NVIDIA's H100 GPU, with a cost-performance improvement of 30-40%. Trainium 3 is scheduled for release in the second half of 2025, with a 2x increase in computing performance and a 40% improvement in energy efficiency.

However, the report also points out that NVIDIA maintains competitiveness in TCO calculations due to its more mature system integration capabilities and stronger software ecosystem, especially in scenarios that require flexibility to respond to different AI tasks.

The research report mentions that the potential rise of quantum computing may impact AI semiconductor demand, but currently, the applicability of quantum computing in AI inference is low, making it difficult to replace ASICs and GPUs in the short term In addition, retired GPUs may pose a threat to the ASIC market. Some cloud service providers may choose to reduce costs by using retired GPUs instead of investing in expensive ASICs.

Morgan Stanley concluded that the rise of ASICs does not mean the decline of GPUs. On the contrary, both technologies will coexist in the long term, providing optimal solutions for different demand scenarios.

In the future AI market, ASICs will strive for a larger share due to their cost and energy efficiency advantages, while NVIDIA will continue to consolidate its market position by relying on its technological leadership