AMD Stock Has Doubled This Year. Is It Still a Buy?

AMD’s data center revenue rose 57% year over year to $5.78 billion in Q1, while Client & Gaming increased 23% to $3.6 billion and Embedded grew 6% to $873 million. The article argues that despite strong AI-driven growth, AMD now looks expensive at 139x trailing earnings and 58x forward earnings versus Nvidia at 27x forward earnings. Overall, the piece is a valuation-focused comparison that favors Nvidia over AMD for new money.

Analysis

The key second-order shift is not simply that AMD is “good at AI,” but that its mix is becoming a levered bet on a single end market just as capital intensity in the AI stack is rising. That creates a paradox: stronger data center momentum can improve near-term growth optics while simultaneously compressing the margin of safety investors once got from AMD’s broader exposure to client, gaming, and embedded. In practice, this means AMD’s multiple is now tied less to semiconductor cycle normalization and more to whether AI infrastructure spend keeps outrunning depreciation and competitive supply additions.

The more interesting relative-value angle is that the market is paying AMD like an accelerating winner while still assigning it a discount to NVDA on growth quality and earnings visibility. That spread is justified if NVDA’s ecosystem lock-in remains intact; if not, AMD’s upside is mostly a function of continued share gains and execution, not valuation re-rating. The risk is that AMD’s recent rerating already embeds much of the “catch-up” narrative, leaving little room for a miss in shipments, ASPs, or guide that would otherwise be tolerated in a less expensive name.

A less obvious beneficiary of the AMD-versus-NVDA debate is the broader AI supply chain: if AMD keeps taking share, demand diversifies away from a single buyer and may improve pricing power for adjacent infrastructure vendors, but it also increases the chance of component bottlenecks and lead-time volatility across memory, packaging, and networking. Conversely, if AMD underdelivers, capital likely rotates back into NVDA rather than out of AI entirely, making this more of a winner-take-more setup than a sector-wide unwind. The contrarian read is that the market may be underestimating how quickly a still-expensive “second choice” can de-rate if AI capex growth merely normalizes from extraordinary levels.

Time horizon matters: over the next 1-3 months, the stock is vulnerable to multiple compression on any guidance reset because the setup is valuation-led, not just earnings-led. Over 12-24 months, the real question is whether AMD can sustain share gains without becoming dependent on a single hyperscaler budget cycle; if not, volatility should remain structurally elevated even in a strong fundamental tape.

Key Decisions for Investors

• Prefer a relative-value long NVDA / short AMD pair over the next 1-3 months: the trade isolates quality-vs-catch-up valuation and benefits if AI spend stays strong but the market re-anchors to earnings durability.
• If already long AMD, hedge with near-dated put spreads into the next earnings/guidance window; the stock’s rerating leaves limited cushion if management merely meets, rather than beats, elevated expectations.
• For new exposure to AI semis, wait for a 10-15% pullback in AMD before adding; the risk/reward is better on valuation resets than chasing momentum after a near-double.
• Monitor NVDA/AMD relative performance as a signal for AI breadth: sustained NVDA leadership suggests the market still prefers ecosystem moat over share-take stories, while AMD outperformance would indicate a broader tolerance for second-source diversification.
• Avoid outright short AMD unless data center growth visibly decelerates for 1-2 quarters; the better expression is short AMD versus long a diversified semiconductor basket, limiting thesis risk to valuation rather than sector beta.