AMD enters its second-quarter earnings report with one number standing above the rest: $6.5 billion in expected data center revenue. That figure has become the clearest test of whether the company’s AI-driven growth story is still accelerating after a volatile stretch for semiconductor stocks.
Shares of Advanced Micro Devices closed at $476.15, down 1.9% in the latest session and about 18.6% below the record high of $584.73 reached on June 30. Even after that retreat, the stock remains up 194% over the past 12 months, underscoring how much future AI demand is already reflected in the valuation.
With AMD reporting after the market close on Tuesday, the focus is not only on whether it beats quarterly estimates, but also on how clearly management can quantify AI GPU demand, customer deployments, and the revenue path tied to its expanding data center pipeline.
Key Facts
- Wall Street expects AMD to report adjusted earnings of about $1.61 per share on revenue of $11.31 billion.
- Data center revenue is projected at $6.5 billion, up from $3.2 billion in the year-earlier quarter.
- AMD has disclosed 14 gigawatts of prospective AI infrastructure commitments across OpenAI, Meta, and Anthropic.
- OpenAI and Meta each received warrants for up to 160 million AMD shares, creating potential dilution of 320 million shares in total.
- Options pricing implies an approximately 4.32% move in AMD shares immediately after earnings.
AMD Q2 earnings preview
The setup into earnings is unusually complex because AMD sits at the center of two very different semiconductor narratives. On one side, it is a beneficiary of AI infrastructure spending through its Instinct accelerators, EPYC server CPUs, networking products, and rack-scale systems. On the other, it still has meaningful exposure to the PC market, where rising memory costs could pressure demand in coming quarters.
Consensus estimates point to strong headline growth. Revenue of roughly $11.3 billion would represent a year-over-year increase of about 47% from $7.6 billion, while earnings per share would jump from $0.48 to roughly $1.61. AMD’s own prior guidance called for around $11.2 billion in revenue and a gross margin near 56%, meaning the market has limited patience for anything that merely matches expectations.
The segment mix explains why this report matters so much. Data center is expected to contribute about 57% of total revenue, up sharply from roughly 42% a year earlier. Client revenue is projected around $3 billion, while gaming is expected near $781 million and continues to shrink. That leaves AMD increasingly dependent on AI and server momentum to justify both its growth multiple and its position in one of the market’s most crowded trades.
AMD does not just need a quarterly beat; it needs to prove that its AI revenue pipeline is becoming visible, scalable, and durable.
Why the 14-gigawatt pipeline matters
AMD has disclosed prospective AI infrastructure commitments totaling 14 gigawatts across three major customers. OpenAI signed a 6-gigawatt agreement in October 2025, Meta announced a separate 6-gigawatt commitment in February 2026, and Anthropic added 2 gigawatts on July 22 tied to MI450 Series GPUs deployed in Helios rack-scale systems.
These commitments are strategically important because they suggest AMD is moving beyond one-off chip sales and deeper into large-scale platform deployments. Helios combines Instinct GPUs, EPYC Venice CPUs, Pensando networking, and ROCm software into an integrated rack offering. With pricing reported in a range of $5 million to $5.5 million per rack, AMD is positioning itself less as a component supplier and more as a systems vendor, a shift that could support larger deal sizes and potentially stronger margins over time.
Still, investors should separate disclosed commitments from booked revenue. Management has identified the usual execution risks, including export controls, tariffs, manufacturing capacity, component supply, product performance, and customer demand. The pipeline is substantial, but the timing of conversion into recognized sales remains one of the biggest variables in AMD’s valuation.
AI growth, CPU strength, and dilution risk
One reason AMD’s investment case has broadened is that the company is no longer viewed solely through the lens of AI accelerators. The EPYC server CPU business has become increasingly important as AI workloads require substantial compute orchestration beyond GPUs. Agentic AI systems, which perform multi-step tasks and manage memory and workflow logic, are reinforcing demand for server CPUs alongside accelerators.
That dynamic matters because it gives AMD a second high-value path to data center growth. If GPU adoption progresses more slowly than bulls expect, EPYC can still help sustain earnings power. The market is likely to press management for more clarity on how much of the expected $6.5 billion in data center revenue comes from GPUs versus CPUs, since AMD does not provide a clean breakout between the two.
At the same time, shareholders have to reckon with the cost of winning hyperscale business. AMD granted OpenAI and Meta warrants allowing each customer to purchase up to 160 million shares at $0.01 per share, for a combined potential dilution of 320 million shares. Those warrants vest in tranches tied to deployment milestones and share-price thresholds through October 2030. In practical terms, dilution increases as the strategy succeeds, which complicates per-share valuation even if underlying revenue growth remains strong.
Implications for Investors
For investors, the immediate question is whether AMD can deliver enough evidence to support expectations already embedded in the stock. A strong quarter by itself may not be enough. The market will likely demand a data center result above or clearly in line with the $6.5 billion target, along with more precise commentary on MI450 shipments, Helios backlog, and revenue visibility into late 2026 and 2027.
The main risk is that AMD posts solid reported results but offers cautious guidance on the client segment. Rising DRAM and NAND pricing could pressure PC demand as manufacturers absorb higher costs or reduce production plans. Since client revenue is still expected to rise about 20% year over year in the current quarter, any memory-related weakness would probably show up more in forward guidance than in the just-ended period.
Valuation also remains a central issue. The average analyst target sits near $576.55, implying meaningful upside from current levels, but target ranges are unusually wide. That reflects uncertainty over how quickly AMD can convert its AI commitments into revenue, how profitable rack-scale systems become, and how much future dilution investors should price in from the warrant structures. With options implying a roughly 4.32% post-earnings move, short-term volatility should be expected regardless of the headline numbers.
AMD’s report is likely to shape sentiment across the semiconductor sector for the rest of the quarter. If management can pair strong data center growth with credible visibility on AI deployments and manageable client risks, the stock could rebuild momentum quickly; if not, investors may keep demanding harder evidence before paying up for the next leg of the AI trade.