KEY POINTS
- Intel posted Q1 2026 revenue of $13.6 billion, up 7% year over year, with Data Center and AI revenue jumping 22% to $5.1 billion and segment operating margin hitting 31%.
- Surging demand for host CPUs in agentic AI systems — including Intel Xeon 6 being selected for Nvidia's DGX Rubin NVL8 — is re-pricing a business that Wall Street had written off two years ago.
- Traders should watch INTC's reaction to next Friday's PCE print and the Q2 revenue guide of $13.8 to $14.8 billion, well above the $13.07 billion consensus.
Intel stock surged as much as 15% in extended trading Thursday after the chipmaker reported first-quarter revenue of $13.6 billion, up 7% year over year, with Data Center and AI revenue climbing 22% to $5.1 billion and segment operating profit hitting $1.5 billion. The print marks the sixth consecutive earnings beat for CEO Lip-Bu Tan and, more importantly, confirms a structural re-rating of Intel's core business that few analysts were modeling six months ago. Non-GAAP operating margin expanded to 12.3% from 5.4% in the year-ago quarter. Intel also guided Q2 revenue to $13.8 to $14.8 billion against a $13.07 billion consensus, and 20 cents of adjusted EPS against 9 cents expected.
The CPU Comeback
For the better part of two years, the story in AI silicon was Nvidia, Nvidia, and some Nvidia. Intel was the company that had missed foundry, missed mobile, missed the first AI wave, and was stuck selling commodity CPUs into a data center market being eaten by GPUs. That narrative is breaking. AI-related businesses now constitute 60% of Intel's revenue and grew 40% year over year in Q1, according to the company's own release. The driver is not a reversal of the GPU boom but a recognition that agentic AI workloads — the orchestration layer, the memory handling, the inference routing — sit on CPUs. When an enterprise deploys a fleet of autonomous agents, the host CPU is doing more work, not less.
That shift showed up in the most consequential disclosure on the call: Intel Xeon 6 was selected as the host CPU for Nvidia's DGX Rubin NVL8 systems. Nvidia's own rack-scale reference architecture for its next-generation GB300-class deployments now runs on Intel silicon. Paired with the multiyear Google collaboration announced last month — involving Xeon processors and custom ASICs for Google's hyperscale estate — Intel has quietly inserted itself into both of the largest AI infrastructure pipelines in the market.
What the Numbers Actually Say
Strip away the AI narrative and the raw financial story is still the cleanest quarter Intel has produced in five years. Gross margin on a non-GAAP basis came in at 38.5%, up more than 600 basis points year over year. The foundry segment, long a drag, narrowed losses materially as 18A-node yields improved into production. Client computing held in at $7.6 billion against expectations of continued PC demand softness. The mix is what matters: high-margin data center revenue is now growing twice as fast as the low-margin consumer business, which flips the earnings algorithm in Intel's favor for the first time since 2021.
Short interest coming into the print was around 4.8% of float. That gets unwound in a tape like this, which partly explains the 15% after-hours move. The more durable question is what happens to consensus numbers heading into the second half. If the Q2 guide is the new baseline and Intel can hold Data Center and AI at 20%-plus growth, full-year 2026 EPS estimates that currently sit in the low 70-cent range need to move toward a dollar. That is a different stock.
Where the Trade Gets Interesting
INTC closed Thursday at $37.48 before the after-hours move. The 52-week high is $39.62, set in early April. A gap open above $42 on Friday puts the stock in price-discovery territory, with the next meaningful chart level near $48 — the May 2020 high. Options traders were pricing roughly an 8% implied move into the print, so the actual reaction is running nearly double what was expected, which typically forces systematic re-hedging from dealers through the first hour of trading.
The read-across matters too. Taiwan Semiconductor ([TSMC reported nearly 41% revenue growth](https://www.stocktitan.net/sec-filings/INTC/8-k-intel-corp-reports-material-event-d8930f0603bf.html) in its most recent quarter, driven by HPC) is the obvious beneficiary as a foundry partner. But the more interesting pair trade is long Intel against the GPU-concentrated names that have already discounted the AI buildout. If Intel's CPU growth story is real, it comes in part at the expense of custom ASIC narratives at Broadcom and Marvell. The AI capex pie is growing, but the slicing is changing.
The Forward Look
The next test is the May 30 PCE report, which will set Fed expectations heading into the June FOMC. Intel's guidance implicitly assumes stable enterprise IT spending through the back half of 2026. If core PCE prints hot and rate-cut expectations get pushed out again, the entire AI infrastructure complex re-prices. Watch the $42 level on INTC — that is where today's move meets resistance from 2022 shareholders who are finally at breakeven and may want out. Above $42, the stock runs. Below, this becomes a grind higher.
