Buy now, ask questions later: CPU stocks soar on AMD’s blowout results, guidance

AMD forecast second-quarter revenue above Wall Street expectations and said the server CPU addressable market could grow over 35% annually to more than $120 billion by 2030, sharply above its prior 18% estimate. Shares rose 17% in premarket trading, while Intel gained 6.2% and ARM rose 11.3% ahead of earnings, reflecting a broader re-rating of the CPU/AI infrastructure complex. Analysts see AMD as a major beneficiary of enterprise agentic AI adoption and potential GPU customer wins, with BofA estimating about 50% CPU market share.

Analysis

The key takeaway is not simply that AMD is taking share, but that the CPU TAM is being re-rated upward because agentic AI increases compute intensity at the enterprise edge of the stack. That is a much better setup for x86 than for “pure AI” narratives: the more AI agents interact with legacy enterprise systems, the more installed-base compatibility becomes a moat, which supports AMD’s share gains while limiting the degree to which ARM can displace x86 in near-term workloads. The first-order winner is AMD, but the second-order winner is the entire CPU ecosystem because a bigger market expands the pie faster than the share shifts can compress it.

Intel’s move likely reflects positioning around a less-bad cycle rather than a structural turn in fundamentals. If the market starts to believe server CPU growth is >35% annually, Intel’s downside is partially cushioned by beta to the category, but the company still faces a widening execution gap if AMD continues to outgrow the market. The more interesting hidden beneficiary may be the infrastructure layer around enterprise AI deployment: OEMs, motherboard/board-level suppliers, and datacenter capex enablers should see a longer runway if CPU demand is tied to AI-agent rollouts rather than a one-time GPU buildout.

The contrarian risk is that consensus may be extrapolating TAM expansion too aggressively before the adoption curve is proven in revenue. Agentic AI can be real and still translate into slower near-term monetization if enterprises throttle spend, optimize workloads, or delay broad deployment until software standards mature. Also, if ARM keeps rallying on narrative alone, there is risk of a crowded trade where the market has priced in meaningful displacement despite the article’s own implication that x86 remains sticky; that makes ARM vulnerable to a post-earnings fade if commentary is merely incremental.

Near term, the trade is about momentum and revisions; over the next 3-6 months, the real catalyst is whether AMD’s AI day converts TAM rhetoric into customer-specific design wins and margin implications. If AMD can name incremental hyperscaler or sovereign GPU demand for 2027+, that supports a second leg up in estimates; if not, the stock may still outperform, but the multiple expansion could stall. Watch for broadening performance into suppliers and for any sign that Intel’s gain is just a relief rally rather than a durable inflection.

Key Decisions for Investors

• Stay long AMD into AI day / next 4-8 weeks; use pullbacks as entry. Risk/reward remains favorable as upward estimate revisions can compound the guide-up, but trim if management fails to quantify incremental enterprise design wins.
• Pair long AMD / short INTC for a 1-3 month horizon. This isolates share-take and execution dispersion; the spread should widen if server CPU TAM revision is real, but cover if Intel shows credible roadmap/customer traction.
• Take a tactical short or put spread on ARM into earnings if the pre-earnings move extends materially. The setup is vulnerable to a 'good but not great' print because expectations have been levered by the broader AI CPU enthusiasm.
• Add selectively to semiconductor infrastructure names on any weakness (board-level, networking, datacenter capex proxies) over the next 1-2 quarters. If enterprise agentic AI becomes a capex wave rather than a narrative, second-order beneficiaries may outperform pure compute beta.
• For longer-term investors, use AMD calls rather than outright stock into July AI day to capture upside convexity while limiting downside if the TAM expansion story proves too optimistic.