AMD stock jumps 7% on earnings as data center revenue nearly doubles

AMD reported Q3 revenue of $13.38 billion, up 45% year over year and above the $13.09 billion LSEG estimate; adjusted EPS was $2.09 versus $1.95 expected, and data-center sales nearly doubled. The company forecast Q4 revenue of about $17 billion, ahead of the $16.2 billion analyst estimate, while projecting a 55% gross margin, slightly below the 55.3% estimate; shares rose 7% in extended trading. PC and gaming revenue were weaker: the combined segment fell 8% to $3.7 billion, with gaming down 40% year over year.

Analysis

The more durable signal is workload breadth: agentic systems may add sustained server-CPU demand alongside accelerator spending, giving AMD a distinct growth lane rather than requiring it to displace NVIDIA in GPUs. The key test is whether that demand converts into repeatable server-CPU revenue and profit, not simply whether AI infrastructure budgets keep rising. The below-consensus gross-margin outlook is a useful counterweight to the growth narrative: monitor whether product mix, ramp costs, or pricing pressure prevent revenue growth from translating into operating leverage.

AMD’s customer equity investments create a second-order risk: they may deepen strategic adoption, but also blur the line between independently funded demand and demand supported by vendor capital. Verify customer deployment volumes and AMD’s cash commitments before treating announced relationships as durable economics. Supply remains another gating variable; the stated plan to raise supply in 2027 is not evidence that near-term constraints are resolved. TSMC and memory suppliers could benefit if deployments scale, but capacity allocation and component availability may limit the upside.

Near term, the earnings/guidance surprise can support the shares, but the large year-to-date run-up makes the bar for follow-through high. Over 1–3 months, watch data-center growth, gross-margin delivery, and customer deployment evidence. Over 6–18 months, PC weakness and gaming declines remain an offset, while the structural thesis depends on agentic workloads becoming material. Intel’s supply and guidance credibility issues may create share opportunities for AMD, but a broader AI-capex retrenchment would hurt both.

Key Decisions for Investors

• Prefer a staged long AMD / short Intel pair over chasing AMD outright: initiate only after the post-earnings move stabilizes, with position sizing tied to relative volatility. The thesis is relative execution and potential CPU share capture, not immunity to an AI-capex downturn.
• Falsify the pair if AMD’s data-center growth decelerates materially, gross margin misses or remains pressured, or Intel demonstrates reliable supply and improving guidance; reassess after the next earnings updates.
• Treat TSMC and memory suppliers as conditional second-order beneficiaries, not direct buys on this report alone. Track capacity allocation, component availability, and confirmed customer deployments before adding exposure.
• Do not underwrite the customer investments as proof of demand. Verify the timing and terms of AMD’s commitments, customer purchase volumes, and whether deployments persist without additional vendor financing.