AMD's Server-Silicon Dominance and the Samsung Pivot: A Two-Pronged Strategy for the AI Era

AMD's server revenue share hits 46.2% as agentic AI drives demand. The chipmaker negotiates with Samsung for 2nm chips to hedge TSMC reliance, while stock soars 72% in 30 days.

AMD has muscled past a historic milestone in the server market while simultaneously negotiating a production lifeline with Samsung Foundry. The chipmaker captured 46.2% of all x86 server revenue in the first quarter of 2026, pushing Intel’s share to just under 54% — a level the smaller rival hasn’t held in decades. The coup came as server-processor shipments jumped 6% quarter over quarter, breaking the typical seasonal decline, with AMD’s own unit sales rising 15% while Intel’s slipped.

Behind the surge is the insatiable appetite of agentic AI workloads. These applications demand processors with a high core count to run multiple AI agents in parallel on a single chip, and AMD’s architecture is winning the architecture race. The company’s data-center revenue rocketed 57% to $5.8 billion in the latest period, part of a total revenue haul of $10.25 billion that beat analyst estimates of $9.89 billion. Management guided for roughly $11.2 billion in the current quarter.

Samsung Talks Signal a Supply-Chain Hedge

To keep the engine running, AMD is in active discussions with Samsung Foundry about producing cutting-edge 2-nanometer chips. Until now, the company has relied almost exclusively on TSMC for manufacturing. With the Taiwanese giant’s capacity largely booked for years ahead, a deal with Samsung would give AMD much-needed production flexibility and reduce its geopolitical exposure to a single island.

The stock has rewarded this expansionary narrative, trading recently at €384.00 — roughly double its level at the start of the year. Yet the valuation remains the subject of fierce debate. The trailing price-to-earnings multiple sits at 153, though forward estimates compress that to 65 as earnings accelerate.

Analyst Targets Converge — but Sentiment Splits

On May 13, two houses landed on sharply different conclusions. Bank of America’s Vivek Arya raised his price target to $500 from $450 and reiterated a Buy, arguing that the total addressable market for AI data centers could reach $1.7 trillion by 2030. Daiwa, meanwhile, downgraded AMD from Buy to Outperform — but simultaneously lifted its target from $250 to $500. The downgrade was driven not by fundamentals but by the speed of the rally: the stock had climbed 72.53% in just 30 days. Daiwa expects earnings per share of $12.85 by 2027, making the new target roughly 39 times that forecast.

Mizuho had weighed in a day earlier, boosting its target to $515 from $415 on the thesis that agentic AI requires both server CPUs and accelerators — a dual demand that AMD is uniquely positioned to serve.

CEO Lisa Su has told investors the server-CPU market could grow at over 70% annually and reach more than $120 billion by 2030. The company’s free cash flow soared 253% year over year to $2.566 billion, underscoring the operational momentum.

The Valuation Elephant in the Room

Even bulls concede the stock leaves little margin for error. The price-to-earnings ratio stands at 146.98, and after the recent run-up, any whiff of disappointment could trigger profit-taking. Insider sales over the past three months amounted to $54.6 million — not necessarily a red flag, but a potential drag on sentiment.

Morningstar analyst Colello also points to in-house chip designs from Google, AWS, Microsoft, and Meta, as well as the Nasdaq debut of Cerebras on May 14 under ticker CBRS. While Cerebras targets training and inference workloads and could nibble at AMD’s market, it is seen more as a niche specialist than a broad platform rival.

The consensus analyst target sits at $395.65, with a wide range from $248 to $530 — a reflection of the deep uncertainty in a market that rewards AMD’s execution but punishes its multiple. The next catalyst will be the second-quarter report: hit the $11.2 billion revenue target and the growth thesis gains fresh oxygen; miss it and the valuation argument becomes far sharper than any competitive threat.