Intel’s first-quarter report was the kind of print that jolts a stock out of its old identity. Adjusted EPS came in at $0.29 versus $0.01 expected, CNBC reported on April 23, and revenue reached $13.58 billion against $12.42 billion expected. Shares jumped 20% in after-hours trading, and by the next session they were changing hands around $81.41, with the day’s range running from $79.62 to $85.22, Investing.com showed.

That move was not a surprise in the abstract. The stock had already climbed 52% in the month before earnings, Morningstar noted on April 24, so the setup was stretched before a single number hit the tape. The market was already leaning forward. The quarter just gave it a shove.

The key point, though, is not the pop itself. It is what the quarter says about Intel’s operating business. Revenue rose 7% year over year to $13.6 billion, Morningstar said, and gross margin reached 41%, roughly 650 basis points above expectations. Q2 guidance was also well ahead of the Street, with Intel forecasting revenue of $13.8 billion to $14.8 billion and adjusted EPS of 20 cents, CNBC reported.

That is the part bulls can lean on. The quarter was not just one clean beat. It was a beat in revenue, a beat in margin, and a forward guide that suggested the strength was not a one-off.

The most persuasive evidence came from Intel’s data center business. Revenue there climbed 22% to $5.1 billion, CNBC reported, and operating margin expanded from 13.9% to about 30.5%, a gain of more than 1,600 basis points, Investing.com calculated. That is the kind of improvement that can change how investors think about a business, because it shows more than just demand. It shows use.

Demand, in this case, seems to be coming from AI-related server CPUs rather than the GPU race that grabs most of the headlines. Morningstar said server processors were in high demand thanks to the rise of agentic AI, and that Intel missed out on at least $1 billion in sales because it could not fully meet demand, Morningstar reported. That is a supply problem, not a demand problem. Investors should not romanticize it, but it is a healthier problem to have.

Intel also picked up some concrete validation. Xeon 6 was selected as the host CPU for Nvidia’s DGX Rubin NVL8 systems, and Google committed to using multiple generations of Intel CPUs for AI workloads, CNBC reported. Those are real names, not vague “ecosystem momentum” language designed to sound comforting in a conference room.

Still, the quarter did not solve every question. Management reported that server CPU average selling prices rose 10% to 15% during the period, Investing.com showed, which helped margins, and favorable product mix also played a role. That leaves an obvious follow-up: how much of 41% gross margin is sustainable once supply constraints ease and the mix normalizes? The quarter gives evidence, not final proof.

This is where the stock gets tricky. Intel is no longer priced like a tired legacy chipmaker waiting for a nostalgia dividend. At roughly $81 to $85, it trades at about 117 times forward earnings, Investing.com calculated. That is a demanding multiple for any company, and especially one that is still carrying losses.

Morningstar raised its fair value estimate to $60 from $32 after the earnings report, Morningstar said on April 24. That is a serious upgrade. It also leaves the shares trading well above that estimate. By Morningstar’s own read, the stock now asks investors to pay ahead of the evidence.

The market is not just pricing Intel’s current business. It is pricing a full foundry transformation, continued strength in AI-driven data center chips, and a packaging business that could scale into something meaningful. Those are three separate bets. Only one of them, the operating turnaround in data center and margins, has clear near-term evidence behind it.

The foundry numbers make the gap obvious. Intel’s foundry revenue reached $5.4 billion in the quarter, but only $174 million came from external customers, CNBC reported. The segment posted a $2.4 billion operating loss in the quarter, and free cash flow remained roughly negative $2 billion, Investing.com reported. That is a long way from a self-sustaining foundry business.

Debt is part of the same story. Intel repurchased the $14.2 billion Fab 34 stake in Ireland during the quarter, took on about $6.5 billion in new debt, and pushed total corporate debt toward $50 billion, Investing.com said. The stock is not just reflecting optimism. It is carrying financing and execution risk in the open.

If there is a second pillar to the bull case, it is advanced packaging. CFO David Zinsner told CNBC that he now expects packaging to bring in billions of dollars per customer, a step up from his earlier estimate in the hundreds of millions. That is a big change in ambition. It also tells you how central packaging has become to Intel’s pitch.

BEP Research’s February analysis argued that Intel’s EMIB technology, introduced in 2017, has already shipped more than a million units across products including Stratix FPGAs, Sapphire Rapids server CPUs, and Ponte Vecchio, BEP Research wrote. The same analysis said Intel’s own data shows EMIB at 90% wafer utilization versus 60% for traditional silicon interposers like CoWoS, BEP Research wrote. If packaging shortages remain tight, that is at least a plausible opening.

The broader market backdrop helps the case. Lead times for AI servers exceed 50 weeks, BEP Research wrote, and Intel has a chance to sell into customers who need capacity more than they need marketing poetry. The Terafab project in Austin adds another layer, with Intel joining Musk’s effort to help “design, fabricate, and package ultra-high-performance chips at scale” for SpaceX, xAI and Tesla, CNBC reported.

But this is also where the skepticism earns its keep. BEP Research noted that many specific customer-win claims still need independent verification, and it flagged major external foundry customer announcements by mid-2026 as the critical test, BEP Research warned. CNBC also reported that Intel remains the only major customer of its 18A fabs, while the Ohio fab has been delayed until 2030, CNBC reported. The packaging story may be credible. It is not yet commercial proof.

This is the cleanest way to separate the debate. Intel can be a stronger company without being a clean buy at the current price.

For the stock to look attractive from here, three things probably need to happen. First, the data center improvement has to keep going, not just for one quarter but for several. The margin expansion and customer wins need to look repeatable, not lucky. Second, the foundry business needs to show external traction, because $174 million in outside revenue is not enough to justify the valuation that now sits on the stock. Third, Intel needs proof that 14A and advanced packaging can attract real customers before patience runs thin.

Morningstar’s caution sits right on that line. The firm said Intel has not yet brought 14A to market, remains heavily exposed to PCs, and still faces stiff competition from AMD and Arm-based vendors in server CPUs, Morningstar wrote. That does not make the bull case wrong. It just means the stock is priced for a much cleaner future than the present one.

The market has already rewarded Intel for better execution. That part is over. The next leg depends on whether management can turn promising operations into durable scale without waiting too long for the foundry side to catch up.

On the business, the answer is increasingly yes. On the stock, it is harder to say that with a straight face at these levels.

Intel’s quarter showed real improvement in revenue, margin, and demand in the data center business. It also showed a foundry operation that remains loss-making and lightly penetrated outside Intel itself, with a valuation that assumes the missing piece arrives on schedule. That is not a small gap. It is the whole argument.

For investors deciding whether to buy Intel stock now, the key question is not whether the turnaround exists. It does. The question is whether the current price already assumes too much of the next phase, before Intel has proved it can scale external foundry and packaging revenue into something durable. Until that happens, Intel looks more like a transformed company than a straightforward buy.