To make the thesis more useful, I use three valuation paths as reference: a downside scenario with a target price of $465.83, a base scenario with $811.45, and an upside scenario with $1,111.89.
Current price at the time this analysis was written: $355.26
🟢 Upside scenario
Here I assume the market is underestimating the speed of AMD’s AI ramp.
In this case, EPYC keeps accelerating more than expected, Helios executes without major hiccups, MI450 gains more traction than anticipated, and the visibility with large customers like Meta and OpenAI translates into much larger and faster revenue.
What makes this scenario plausible is that AMD is not just talking about demand, but about lead customer forecasts that are already above initial plans for MI450, with Meta expanding to as much as 6 gigawatts and new customers showing interest in multi-gigawatt deployments.
I also think it matters that AMD raised the server CPU TAM to more than $120B by 2030, because that expands the opportunity set and makes the growth assumption less demanding if AI adoption keeps scaling.
If the server CPU mix keeps pushing margins while data center AI scales, AMD could justify another multiple expansion.
In this scenario, I would see a reasonable valuation range of $811.45 to $1,111.89 per share, with $1,111.89 as the aggressive midpoint reference.
That would imply an upside of about 128.4% in the midpoint case and 213.0% in the most optimistic case, versus the current price of $355.26.
🟠 Base scenario
My base case is that AMD keeps executing well, but without requiring perfection. I assume the server CPU business maintains a very strong pace, Instinct/Helios begins contributing more materially in the second half, and the company continues gaining share in data center without the market handing it an unlimited multiple expansion.
What supports this scenario most is that the company already showed real traction: Q1 revenue of $10.3B, up 38% YoY, with data center at $5.8B (+57%), and Q2 guidance of $11.2B ± $300M, which implies +46% YoY at the midpoint.
I also care that AMD said server CPU business would grow more than 70% YoY in Q2, with strong continuity into the second half of 2026 and into 2027, because that tells me the engine is nowhere near stalling.
In addition, the company reported record free cash flow of $2.6B, gross margin of 55%, and guided gross margin of 56% for Q2, which suggests the business is growing on a much sturdier financial base than the market often assumes.
In this scenario, I would see a reasonable valuation range of $465.83 to $811.45 per share.
My midpoint would be $811.45. That would imply an upside of about 128.4% versus the current price of $355.26.
🔴 Downside scenario
In the downside case, I assume the market runs into two limits: first, a more complicated AI ramp than expected; second, stronger margin and demand pressure from memory, the supply chain, and implementation costs.
I am not saying the thesis breaks, but I am saying it becomes less linear and slower to monetize.
The call itself points to the risks: AMD acknowledged tightness in the supply chain, memory pressure, and pressure in PCs and gaming from higher memory and component costs.
AMD also expects gaming to fall more than 20% in the second half versus the first half, which removes part of the mix support.
And while the company sounded confident about the Helios ramp, the fact that MI450 starts scaling in Q3 and accelerates in Q4 means we are still at a stage where execution can materially move the outcome.
In this scenario, I would see a more conservative valuation range of $355.26 to $465.83 per share, with $465.83 as the reasonable ceiling if the thesis cools but does not break.
That would imply an upside of about 31.1% in the best case within this downside scenario, versus the current price of $355.26.
This analysis represents a personal opinion based on a review of the company’s public reports and does not constitute investment advice.
After this report, my position does not change in the essentials: I still see AMD as a structurally stronger story than the market used to recognize.
What does change is the scale of that story. This report makes me think the thesis no longer depends only on AMD “gaining share”; it depends on AMD building a much larger data center business, with better visibility, more CPUs per AI deployment, and a second accelerator leg that is starting to take real commercial shape.
For that reason, my base scenario stays constructive, my upside scenario becomes more credible, and the only reason to become more cautious is not demand — it is execution of that demand in a supply chain that is still complex.