Financial Report Preview | AI chip duel war upgraded! AMD (AMD.US) earnings report will test Nvidia's “challenger”

The Zhitong Finance App learned that AMD (AMD.US) will announce its second-quarter earnings report after the US stock market on August 4 (Tuesday). As tech giants continue to increase investment in AI infrastructure, the market is closely watching whether AMD can continue to win orders for AI chips and further narrow the gap with industry leaders. As Nvidia's core computing power challenger, AMD's current financial report will also be a key weather vane for the market to determine whether the global demand for AI computing power can continue.

The market generally expects AMD's second-quarter revenue to be US$11.32 billion, up 47.3% year on year; earnings per share will be 1.61 US dollars, up 235.42% year over year.

According to data from Zacks, AMD's profit for the past four quarters has exceeded market expectations, with an average of 6.5% higher than market expectations.

AMD's stock price growth doubled during the year, and the explosion in the data center business was the core driving force; however, affected by the overall correction in the AI sector, the stock retreated close to 19% from its June high. Based on recent option prices, traders are betting that AMD's stock price will fluctuate as much as 10% this week.

If AI chip revenue and guidance for the second half of the year exceed market expectations, AMD's earnings report may boost the entire semiconductor sector; if order growth, product delivery, or gross margin performance fall short of expectations, it may increase investors' concerns about the return on AI capital expenditure.

What are the highlights of AMD's Q2 earnings report?

Data center business remains a core growth driver

The market generally anticipates that the data center business will continue to be the core growth driver for AMD's second quarter results. Strong demand for EPYC server CPUs and Instinct AI accelerators is expected to drive AMD's revenue growth. Strong demand for AI inference, accelerated implementation of enterprise AI, combined with major cloud vendors continuing to increase investment in AI infrastructure, is expected to boost Instinct GPU shipments. At the same time, the wide application of fifth-generation EPYC processors in hyperscale data centers and enterprise customers is expected to support the revenue of server CPUs.

It is worth noting that the market attention for AMD's next-generation hardware products continues to heat up. AMD CEO Su Zifeng recently stated that overall customer demand for the Mi450 and Helios was higher than initially anticipated, and even surpassed the company's internal plans for 2027. The phenomenon of customers queuing up for deployment shows strong market interest in its platform.

According to AMD's forecast, by 2030, the AI accelerator market will reach 1.4 trillion US dollars, the data center CPU market will reach 220 billion US dollars, and the overall computing market will reach 2 trillion US dollars. The core logic behind this judgment is that the need for AI infrastructure is no longer limited to training large models. As inference and intelligent AI bring new computing requirements, AMD expects server business growth to accelerate further.

Deeply bind the joys and worries of AI giant customers

An important increase in AMD's performance in the second quarter came from large-scale procurement from the world's leading cloud vendors. Mainstream cloud platforms such as AWS, Microsoft Azure, and Google Cloud continue to deploy EPYC processors, and have reached partnerships with Meta, Samsung, Tata Consulting Services, NAVER Cloud, and Upstage, which is expected to drive AMD's AI infrastructure revenue growth.

Among them, Meta clearly stated that it plans to deploy AMD Instinct GPUs and use next-generation EPYC processors, superimposing the continued implementation of global sovereign AI projects, and the excellent performance of AMD products in MLPerf benchmarks, further boosting industry customer confidence and helping AMD win more orders.

From a positive perspective, deep binding to leading AI giants is the core support for AMD's medium- to long-term computing power business growth. Last year, AMD finalized a major strategic partnership to bind OpenAI through a multi-billion dollar share warrant cooperation in exchange for the other party's large-scale procurement of its own MI350X chips. OpenAI's latest operating data further strengthens optimistic expectations. OpenAI recently revealed that its annualized revenue for July has surpassed the sum of the second quarter.

At the risk level, the iteration of the open source model is quietly changing the logic of the industry's computing power requirements. The market is concerned that the rapid iteration and popularization of open source models will impact the pricing and customer base markets of closed source vendors such as OpenAI and Anthropic. More importantly, the open source model is lightweight and efficient, and the GPU computing power required to complete the same AI tasks is drastically reduced, which may significantly reduce the overall computing power procurement demand of the industry. Customers from closed source manufacturers continue to be lost, and demand for superimposed model computing power has shrunk drastically. In the long run, it will directly reduce the market space for AMD AI chips and weaken the logic of high growth.

PC and embedded businesses are expected to perform steadily

In addition to the booming data center business, AMD's traditional business segment is expected to perform steadily.

Among them, the client business is expected to benefit from stable demand for Ryzen processors and the continued increase in the penetration rate of the AI PC industry. Enterprise equipment updates and iterations, large-scale implementation of commercial Ryzen AI PRO processors, and intensive launch of new Copilot+ AI PC products are expected to boost AMD's desktop and notebook processor shipments.

At the same time, AMD's share in the high-end consumer PC market continued to expand, and demand for high-end Ryzen X3D processors for gaming and creative design scenarios was strong, which also helped drive overall revenue growth in the second quarter.

Furthermore, as demand for scenarios such as industrial automation and edge AI continues to improve, the revenue of AMD's embedded business is likely to remain stable.

Competition in the industry continues to heat up, limiting revenue growth and gross margin recovery

AMD is currently facing fierce competition from Nvidia, Broadcom, and Intel on the core track.

In the field of AI accelerators, Nvidia absolutely dominates the market with the CUDA ecosystem and a market share of up to 80%, but AMD is becoming a challenger that cannot be ignored through full-stack solutions, cost-effective hardware, and key customer breakthroughs.

What is particularly noteworthy is that on July 23, AMD announced a series of new products at the Advanced AI Conference. The company is trying to comprehensively challenge Nvidia through a “full stack” product layout from the chip to the complete machine system.

AMD claims that its Helios platform based on the next-generation MI450 series accelerator can beat Nvidia's next-generation Vera Rubin platform in some key indicators; the performance of the new-generation EPYC “Venice” server CPU is about 20% higher than that of Nvidia's competitor products. AMD also said that Helios is “in full production” and that “customer demand for Helios is extremely strong” and is expected to begin shipping at the end of the third quarter, and Venice-based server systems are expected to be shipped one after another in the fourth quarter of this year.

Furthermore, Intel is actively competing for server market share through an expanding Xeon processor product line, foundry business, and advanced packaging technology. Broadcom continues to deeply cultivate the hyperscale cloud customer market, and continues to consolidate its competitive advantage in the fields of customized AI accelerators and high-performance network equipment, further squeezing AMD's market space.

Continued intensification of industry competition may affect AMD's prospects for revenue growth and profit margin expansion in the second quarter of 2026.

Wall Street analysts have a long track record

Wall Street analysts are generally optimistic about AMD. Of the 35 analysts followed by Tipranks, 28 recommended buying AMD, 7 gave a “neutral” rating, and the consensus rating was “strong buy”. The average target price was $582.65, which is 22% higher than last Friday's closing price.

The iterative upgrade of AMD's next-generation AI hardware products is the core logic that analysts are optimistic about the company's subsequent growth, and it is also a key observation direction for this financial report.

AMD unveiled a range of new products at the “Promised AI” event held in July. Subsequently, analysts at UBS and Bank of America raised AMD's target share prices from $700 and $560 to $730 and $620, respectively. Analyst William Blair said after the event that AMD “is now actively taking action to seize market share. With a robust product roadmap and focus on open standards, it is expected to win large numbers of orders from hyperscale data centers and artificial intelligence lab customers.”