AMD (NASDAQ:AMD) 20% Down in a Week, Reset Opportunity or Start of a De Rating?

The Sell Off Isn’t About One Quarter, It’s About the Multiple

AMD has sold off hard, down around 21% in the last five trading days after a revenue miss. Even with the share price pulling back toward the $200 level, it feels like there is more going on here than a simple “bad quarter” narrative.

Because on the surface, the quarter was not a disaster at all. In many respects, it actually looked solid.

The issue is expectations. The market had been pricing in explosive AI-driven upside, and in that context, even strong growth can come through as “not strong enough.” When a stock is positioned for perfection, anything short of a blowout can trigger a sharp reset.

So the real question is not just whether AMD executed well. It is whether the reset in expectations has created an opportunity.

After a 20% drawdown, is AMD now a stock worth revisiting, or is this the start of a longer de-rating as AI growth assumptions get recalibrated?

What are the Best ASX Stocks to invest in right now?

Check our buy/sell tips

The Market Wanted a Blowout, It Got Solid, and Punished It

Q4 is one of those quarters where the headline numbers look genuinely strong, even if the market chose to focus on what was not perfect.

Revenue came in at US$10.27 billion, up 34% year on year, and full-year revenue totalled US$34.64 billion, also up 34%.

The mix is telling. The Data Center segment delivered a record US$5.4 billion in Q4 revenue, up 39% year on year, and it has now become the primary growth engine of the company. That growth supports the ongoing AI build-out narrative, plus continued server CPU momentum through EPYC.

But there is a second-order implication investors should not ignore. As Data Center becomes the core driver, AMD is increasingly exposed to the pacing of AI capex cycles, hyperscaler digestion periods, and competitive intensity in accelerators. That dependence can amplify both upside and volatility when expectations are high.

Profitability also moved in the right direction. AMD reported gross margin of 57% in Q4, up from 54% a year earlier, and the improvement reflects both mix and operating leverage.

There is an important nuance here though, because Q4 benefited from two specific items tied to the China export-control situation and inventory accounting.

First, Q4 included an approximately US$360 million release of previously reserved MI308 inventory and related charges, which helped lift reported margins. Second, the quarter included approximately US$390 million of MI308 revenue shipped to China. Both items flattered the quarter versus a clean run-rate view of underlying demand and margin.

Is this the time to buy?

The evidence is still pretty clear that AI underpinned a strong quarter, and the structural demand tailwinds behind data centres have not suddenly disappeared. If you strip away the noise, this feels less like an “AI is broken” moment and more like a pricing and expectations reset.

The real issue is valuation and the way the market has been underwriting the AI opportunity. For the last cycle, investors have effectively priced a wide range of players as if they are all going to be winners. That creates elevated expectations across the board, and AMD has been firmly in that basket. When expectations are stretched, “good” results can still disappoint because the bar is set at “great.”

So the takeaway is not that AMD is failing. It is that the stock is not a bargain.

Even after the drawdown, the share price still reflects a market view that AMD captures meaningful AI upside, sustains strong data centre growth, and keeps expanding margins. If any part of that path looks less certain, the multiple compresses.

In that context, this is not an obvious buy-the-dip setup. The price is still doing a decent job of reflecting the reality of growth and the uncertainty embedded in how big the AI opportunity becomes, and who ultimately captures it.

As we move through 2026, I think we are likely to see that dispersion become clearer.