Prediction: It's Not Too Late to Buy AMD Stock as Revenue Surges

AMD reported Q1 revenue of $10.25 billion, up 38% year over year, with adjusted EPS of $1.37, ahead of the $1.29 consensus. Data center revenue surged 57% to $5.8 billion, while adjusted gross margin improved 170 bps to 55%. Management also lifted its server CPU TAM outlook to $120 billion by 2029 and said data center CPU revenue growth should accelerate to 70% in Q2.

Analysis

AMD’s update matters less as a one-quarter beat than as a signal that the center of gravity in AI infrastructure is broadening from accelerators to the surrounding CPU stack. If the server CPU market really inflects toward the company’s new sizing, the second-order winner is not just AMD revenue growth; it is a higher attach rate of platform-level sockets, memory, and networking content per deployed AI server. That tends to expand the whole bill of materials, but it also shifts bargaining power away from the incumbent CPU leader and into the hands of whichever vendor can bundle roadmap credibility with AI-ready platform performance.

The market is likely underestimating the duration of this cycle because CPU demand is lagging the GPU buildout by a few quarters, not years. That creates a window where AMD can compound share gains while the installed base of AI servers is still being wired, cooled, and upgraded for agentic workloads. The key non-obvious upside is that higher core counts and ASPs can offset any future normalization in unit growth, making the earnings trajectory less dependent on explosive shipment volumes than many investors assume.

The main risk is that consensus may be extrapolating an attractive narrative into 2027 before enterprise deployment behavior is proven. If model architectures shift toward more specialized inference offload or if customers optimize around lower-cost CPU footprints, the TAM expansion could prove less linear than management guidance implies. Near term, the stock can keep working on upgraded estimates, but over 6-12 months the trade becomes increasingly sensitive to any sign of gross-margin compression from mix, competition, or supply constraints.

From a positioning standpoint, this is more compelling as a relative-value expression than as a standalone chase at these levels. The cleanest debate is whether AMD’s share gains are structural enough to compress the premium on incumbent CPU exposure while leaving the GPU leader relatively insulated. If the CPU cycle is real, investors should expect a follow-through in server OEMs, NIC vendors, and memory suppliers, but the biggest disappointment risk sits with names whose valuations still assume durable CPU share stability.

Key Decisions for Investors

• Stay long AMD on pullbacks over a 3-6 month horizon, but use call spreads rather than outright stock to avoid paying up after the move; the setup still supports upside if estimates continue to rise, though asymmetry is less attractive after a 90%+ YTD run.
• Pair trade: long AMD / short INTC for the next 6-12 months; this captures share-shift and roadmap credibility without taking broad semiconductor beta, and the risk is mainly an execution surprise from the short leg rather than macro demand.
• Add a basket long in server-enabling secondaries (e.g., SMCI, MRVL, CRUS or memory exposure) on any post-earnings consolidation; these names can benefit from higher socket complexity and richer content per AI server even if AMD’s multiple stalls.
• For new money, wait for a 10-15% retracement in AMD before initiating size; chasing after a guidance-driven gap increases downside if the market starts discounting 2026-2027 instead of 2025.
• If you already own NVDA, consider reducing overlap rather than adding blindly; AMD’s CPU momentum is a partial hedge to GPU concentration, but the clean relative expression is AMD vs. INTC, not AMD vs. NVDA.