AMD Q1 Preview: I Own It, But I'm Not Adding More Before The Print

AMD is facing a cautious setup ahead of Q1 earnings, with the author refusing to add to a 1.8% NLV position due to the risk of another double-digit selloff. Q4 benefited from one-off China tailwinds, but Q1 MI308 China sales are guided far below Q4 and local chipmakers already hold meaningful market share. The commentary remains long-term bullish, but near-term expectations are clearly restrained.

Analysis

AMD’s near-term setup is a classic “good company, bad tape” trade: the business can still compound over a multi-year horizon, but the next print is likely to force the market to re-underwrite China-related AI revenue at a much lower run-rate. The second-order issue is not just the absolute drop in a single geography; it is the signaling effect on gross margin durability and utilization assumptions, which can compress the multiple even if core demand outside China stays intact.

The more important competitive dynamic is that any shortfall in restricted-node shipments creates a vacuum for local silicon and adjacent systems vendors to lock in design wins, software optimizations, and channel relationships now. That makes the loss partially sticky: once customers retool around alternatives, the revenue can come back slower than the headline guidance cut suggests. In parallel, hyperscale buyers will likely use the uncertainty to press for better pricing across the AI accelerator stack, which could pressure not only AMD but also higher-margin peers and certain packaging/test suppliers.

The risk window is asymmetric into earnings: the market can absorb a gradual slowdown over months, but a guidance reset concentrated in one quarter can trigger a 10-15% air pocket in days if investors conclude the China contribution was masking underlying momentum. What could reverse that is not a vague “beat,” but evidence that ex-China MI demand is accelerating enough to offset restricted shipments, or that mix is shifting toward software/CPU attach where margins are more resilient. Absent that, the stock likely remains supported by long-term AI optionality but capped by skepticism around near-term visibility.

Contrarianly, the consensus may be underestimating how much of the bad news is already embedded after the recent run, especially if positioning has repaired and short interest is not extreme. That argues for tactical downside hedges rather than outright de-risking if one believes the structural AI thesis remains intact. The more interesting mispricing may be in relative value: if AMD sells off on China noise while broader AI capex remains firm, the market could create a better entry point versus semis with more direct export sensitivity or less earnings elasticity.

Key Decisions for Investors

• Do not add to AMD into earnings; if holding a core position, hedge with 1-2 week puts or a put spread into the release to protect against a 10-15% gap-down.
• If AMD sells off >10% on a China guidance reset, look to re-enter selectively for a 3-6 month rebound only if management shows ex-China acceleration; use staged buys rather than a full position.
• Pair trade: long broader AI infrastructure beneficiaries with less China exposure against AMD if the market over-discounts the whole AI complex on one guide-down; prefer names with cleaner U.S./hyperscale exposure and steadier margin mix.
• Watch for a post-earnings rerating to create a relative-value long in AMD versus domestic semiconductor names if the selloff is purely geographies-driven and not a core-demand issue.