📋 Desk Briefing
Intel just delivered its most stunning quarter in over a decade, with revenue growth not seen in 15 years. The results blew past Wall Street expectations on every key metric, driven by a massive 59% surge in its Data Center and AI (DCAI) division. Yet, in the face of this blowout report, Goldman Sachs is holding firm on its ‘Neutral’ rating, creating a fascinating tension for investors.
The core of the debate is whether this quarter marks a genuine, sustainable inflection point or a well-executed beat against lowered expectations. Goldman’s candid response acknowledges the ‘great quarter’ but points to the more attractive risk/reward profiles of Intel’s peers. This analysis dives deep into the numbers, the nuance of the Goldman take, the critical progress in Intel’s foundry ambitions, and the fierce competitive landscape that justifies Wall Street’s lingering caution.
📊 Earnings & Financials
The Q2 2026 numbers were impressive across the board, handily beating analyst estimates. Revenue came in at $16.1 billion, a full 25% higher than the prior year and well above the consensus estimate of around $14.4 billion. Non-GAAP Earnings Per Share (EPS) was $0.42, nearly double the Wall Street forecast of approximately $0.22. This performance marks Intel’s seventh consecutive quarter of exceeding its own guidance, suggesting a management team that is executing consistently.
The headline driver was the Data Center and AI (DCAI) group, which saw revenue explode by 59% year-over-year to $6.3 billion. CEO Lip-Bu Tan noted that demand from cloud and enterprise customers for AI infrastructure is accelerating, to the point where supply is now a constraint. Gross margin also showed significant recovery, expanding to 41.8% (non-GAAP) from just 29.7% a year ago, reflecting better factory utilization and a richer product mix.
However, the Intel Foundry segment remains the key area of scrutiny. While segment revenue grew 31% to $5.8 billion, the vast majority of this is from manufacturing Intel’s own products. External foundry revenue was just $293 million, or about 5% of the total, and the unit still posted a substantial operating loss of $2.1 billion. This is the crux of the caution from firms like Goldman Sachs: while the core business shows a strong cyclical recovery, the transformative foundry story is not yet proven in the financials.
| Metric | Live Data (Q2 2026) | Strategic Read |
|---|---|---|
| Total Revenue | $16.1B (+25% YoY) | Strongest growth in 15 years, beating estimates by over $1.5B. |
| Non-GAAP EPS | $0.42 | Nearly doubled analyst consensus of $0.22. |
| Non-GAAP Gross Margin | 41.8% | Major recovery from 29.7% in the prior year, indicating improved operational efficiency. |
| Data Center & AI (DCAI) Revenue | $6.3B (+59% YoY) | The primary growth engine; demand for Xeon processors in AI roles is outpacing supply. |
| Client Computing Revenue | $8.9B (+13% YoY) | Solid growth suggests the PC market has stabilized and the AI PC cycle is beginning. |
| Intel Foundry Revenue | $5.8B (+31% YoY) | Headline growth is strong, but external customer revenue remains minimal (~$293M). |
| Q3 2026 Revenue Guidance | $15.8B – $16.8B | Midpoint of $16.3B is well above prior consensus, suggesting sustained momentum. |
🔍 Deep Dive: Competitive Moat
Intel’s long-term survival and ambition to reclaim its throne rests on a two-part strategy: regaining process technology leadership and building a world-class foundry business to compete with TSMC. The company’s ‘five nodes in four years’ strategy culminated with the Intel 18A process entering high-volume manufacturing. This is a monumental achievement, bringing groundbreaking technologies like RibbonFET (Gate-All-Around) transistors and PowerVia backside power delivery to mass production for the first time in the industry. The company has already moved on to an enhanced 18A-P variant, which has entered risk production and promises a 9% performance boost.
The critical test is translating this technical progress into external foundry customers. So far, the wins are promising but not yet definitive. Microsoft and Amazon are confirmed partners for the 18A node. However, a blockbuster deal with a major fabless semiconductor company like Qualcomm or AMD remains elusive. Reports in July indicated that NVIDIA tested Intel’s 18A process but decided against a production commitment for now, a significant headline that weighed on the stock.
In the AI accelerator space, Intel is the distant challenger to NVIDIA’s overwhelming dominance. NVIDIA controls an estimated 80-90% of the market for AI data center GPUs. Intel’s offering, the Gaudi 3 accelerator, presents a compelling economic argument. It delivers competitive performance on key AI workloads at roughly half the cost of an NVIDIA H100. While Gaudi 3 won’t beat NVIDIA’s latest Blackwell chips on raw performance, its lower power consumption and open, Ethernet-based networking offer a lower total cost of ownership that could appeal to cost-sensitive enterprises. The challenge is convincing developers to move away from NVIDIA’s deeply entrenched CUDA software ecosystem, a significant moat that locks in customers.
💡 Investment Scenarios
The turnaround under CEO Lip-Bu Tan proves a resounding success. The 18A and 14A process nodes achieve yield and performance parity—or better—with TSMC, attracting major fabless customers who are desperate for a viable second source. Intel Foundry Services breaks even in 2027 and becomes a significant, high-margin revenue contributor. Simultaneously, Gaudi AI accelerators gain meaningful share in the AI inference market, and Xeon CPUs continue to benefit as essential components in all AI infrastructure, validating the company’s multi-faceted AI strategy.
The foundry strategy falters. Despite technical progress, Intel fails to secure high-volume orders from marquee customers, leaving the multi-billion dollar fab investments to be supported primarily by its own products. Process roadmap execution slips, and TSMC maintains its performance lead, keeping customers locked in its ecosystem. AMD continues to chip away at server CPU market share, while NVIDIA’s software moat makes the AI accelerator market impenetrable. Intel ends up with massive capital expenditures and a structurally lower-margin business, causing the stock to de-rate significantly.
🏢 Management Watch
CEO Lip-Bu Tan characterized the quarter as Intel’s “strongest revenue growth in more than fifteen years,” crediting “greater speed, accountability, and customer focus.” His commentary on the earnings call emphasized that the AI buildout is driving unprecedented demand for all forms of compute, especially the x86 CPUs that remain the bedrock of data centers. He noted that demand for Xeon processors is accelerating, with the company currently in a “capacity shortage position.”
A key signal of management’s confidence was the decision to significantly raise capital expenditure guidance. Intel now expects to spend over $20 billion in 2026, up from a previous forecast of $18 billion, with spending set to increase again in 2027. CFO Dave Zinsner framed this as a necessary investment to support the expected growth across both internal products and the burgeoning foundry business. This ramp-up in spending underscores the company’s full commitment to its long-term strategy, even as it weighs on near-term free cash flow.
⚡ Bottom Line
Intel’s stunning quarter proves its core business is recovering strongly, powered by the insatiable demand for AI infrastructure. The operational improvements are real, and the growth in the data center is undeniable. However, Goldman Sachs’s cautious ‘Neutral’ stance is also justified. The stock’s valuation hinges on the long-term, high-risk transformation into a leading-edge foundry, a thesis that remains largely unproven by external customer commitments and will require billions more in investment before its success or failure is clear.
This content is for informational purposes only and does not constitute investment advice. Investment decisions are the reader’s own responsibility. The Scope assumes no legal liability for outcomes.