Nvidia's Controversy: AMD's CEO Lisa Su Praises Jensen Huang, Data Center Business Expected to Account for 80% Next Year

By: www.laohunote.com|11/17/2026 21:53:47

As Nvidia finds itself embroiled in the controversy surrounding Jensen Huang, its long-time rival AMD's CEO Lisa Su has chosen not to kick them while they're down. Instead, she has expressed a focus on AMD's own development and emphasized that the industry's "generational" growth prospects are far beyond traditional cycles.

At the TomorrowX summit held in Austin, Texas, on Tuesday, the 17th, Su publicly praised Nvidia and its recently resigned CEO Jensen Huang, calling Nvidia an "incredible company" and Huang an outstanding leader who has driven the entire industry forward. At the same time, she stressed that AMD needs to shift its attention away from various industry distractions and focus on the tremendous opportunities presented by AI infrastructure development.

Su also provided an optimistic outlook on the current growth prospects of the semiconductor industry. She believes that this round of industry growth is different from traditional cyclical expansions and resembles a "generational" wave of growth. As AI gradually enhances productivity across various sectors, the motivation for companies to invest in AI infrastructure is expected to continue.

On the business front, AMD is accelerating its transformation into a data center company. Su predicts that the data center business will account for over 80% of AMD's total revenue next year, with the CPU market expected to expand to a scale of $220 billion in the next four to five years.

Jensen Huang is an "Outstanding Leader"; AMD Should Focus on Itself and Filter Out the "Noise"

When discussing the recent attention on Nvidia and Jensen Huang, Su did not shy away from addressing her competitor.

She stated, "Nvidia is an incredible company, and Jensen Huang has always been an outstanding leader driving the entire industry forward."

However, she quickly redirected the conversation back to AMD's own strategy.

Su mentioned that many things are happening in the industry right now, but from both AMD's and the industry's perspective, there is a need to some extent to draw attention away from various distractions. "These noises will eventually dissipate over time."

She emphasized that AMD currently has enormous development opportunities, and the key is to participate in this large-scale infrastructure construction, working with partners to advance the future product roadmap and meet or even exceed market expectations.

This statement is quite nuanced. Recently, Nvidia has become the market focus due to Huang's resignation and the possibility of his return as CEO, while Su has chosen to publicly affirm her competitor's industry leadership while emphasizing that AMD should focus on its own business opportunities.

AI Agents Drive CPU Demand; Server CPU Market Expected to Reach $220 Billion by 2030

Another major focus of Su's speech was the profound changes occurring in AMD's business structure.

She pointed out that the share of the data center business in AMD's total revenue is rapidly increasing, from about 25% in 2025 to an expected 60% in 2026, and reaching about 80% in 2027.

This means that AMD is further transforming from a semiconductor company that traditionally relies on personal computers and gaming chips as significant revenue sources to one that centers around data centers and AI infrastructure.

This shift aligns with AMD's recently announced performance and growth guidance.

The financial report shows that AMD's data center business revenue in the third quarter of 2026 grew by 97% year-on-year, reaching a record $8.55 billion. The company expects that by 2027, the data center business will account for over 80% of total revenue, becoming the core driver of overall revenue and profit growth.

Meanwhile, Su particularly emphasized the importance of CPUs in AI infrastructure.

As AI agents gradually take on more complex tasks, computing demands are no longer solely reliant on GPUs for model training and inference. CPUs will also need to handle task coordination, workflow management, and the operating environment for agents.

She predicts that the server CPU market will expand to about $220 billion by 2030. With the development of agent applications, the role of CPUs in AI data centers is expected to further enhance.

This also means that the investment opportunities brought about by AI infrastructure development are not limited to GPUs and accelerators; CPUs may also see significant incremental demand.

AI Programming Capabilities Lower Chip Adaptation Barriers; Anthropic Completes AMD Platform Adaptation in About 3 Days Using Mythos

When discussing the competition between AMD and Nvidia in the AI chip market, Su also touched on the critical issue of the software ecosystem.

For a long time, Nvidia's CUDA software platform and its developer ecosystem have been seen as a crucial part of Nvidia's competitive advantage in GPUs. The accumulated development tools, software libraries, and developer base of CUDA make it difficult for competitors to catch up with Nvidia based solely on hardware performance.

However, as AI model programming capabilities continue to improve, developers' efficiency in generating and optimizing underlying computing kernels using AI is also increasing, which may lower the software adaptation barriers between different chip platforms and challenge traditional software ecosystem barriers.

Su mentioned a case involving Anthropic. Anthropic previously disclosed that its latest model, Claude Mythos Preview, helped the company complete prompt processing adaptation for a model on the AMD Instinct MI355X GPU in about three days, during which nearly 100% of the relevant computing kernels were written or ported independently.

Anthropic stated that using traditional engineering methods, similar work could require dozens of engineers spending 6 to 12 months to complete.

This case shows that AI programming capabilities are changing the efficiency of chip software adaptation and may provide new market opportunities for AMD and other Nvidia competitors. However, this adaptation was specifically for prompt processing of a particular model on AMD GPUs and does not mean that Anthropic has migrated its entire model training and inference business to the AMD platform.

For AMD, if AI can continue to lower software development and porting costs, the long-term accumulated ecological advantages of CUDA may face new challenges, and AMD is likely to further attract AI model developers and enterprise customers.

-- Price

--
--
--

The Growth of the Semiconductor Industry is No Longer a Traditional Cycle but a "Generational" Opportunity

Regarding the growth prospects of the entire semiconductor industry, Su provided a rather optimistic assessment.

She stated that the traditional semiconductor industry has obvious cyclical characteristics: when demand is strong, manufacturers expand production capacity; as supply gradually meets demand, the market then enters a period of inventory digestion and supply-demand adjustment.

However, she believes that the current industry has not exhibited typical cyclical demand characteristics.

"What we are seeing now is not a market that appears to be cyclical, but rather an opportunity that is generational in terms of growth demand."

Su explained that AI will ultimately help every company and every industry improve productivity. Companies are willing to continue investing in AI to develop better products and improve operational efficiency, thereby creating a positive cycle.

She also emphasized that AMD always focuses on the relationship between investment and return, and companies need to ensure that capital investments can yield corresponding returns. However, this current round of growth is indeed different from previous semiconductor cycles.

This assessment also echoes the recent trend of global tech giants continuously increasing investments in AI infrastructure. Despite ongoing discussions in the market about whether AI capital expenditures can be sustained, the return cycles of data center investments, and the sustainability of chip demand, Su believes that the productivity improvements brought about by AI will create long-term growth space for the entire semiconductor industry.

For AMD, this growth wave represents not only competitive opportunities in the GPU market but also the expansion of the server CPU, data center systems, and supporting infrastructure markets. The core message conveyed by Su's speech is that AMD does not need to overly focus on competitors and short-term industry noise; what is more important is to seize the long-term, structural growth opportunities brought about by AI infrastructure development.

This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.

You may also like

iconiconiconiconiconiconiconiconicon
Customer Support:@weikecs
Business Cooperation:@weikecs
Quant Trading & MM:bd@weex.com
VIP Program:support@weex.com