Hook & thesis
Intel just reported a clean operational pivot: revenue of $16.1 billion, adjusted EPS of $0.42, and a Data Center & AI (DCAI) segment that grew 59% year-over-year. The market punished the print because of a large GAAP net loss tied to a non-cash CHIPS Act accounting item and a much-larger-than-expected capex trajectory. That reaction created a tactical buying opportunity for disciplined investors who want exposure to the AI server cycle without paying peak multiples.
Thesis in one line: buy Intel at $85.24 and ride the momentum in DCAI and early foundry customer wins into broader margin recovery - but do it with a firm stop because elevated capex and an unprofitable foundry unit keep headline risk high.
What Intel does and why the market should care
Intel designs and manufactures processors and systems spanning client PCs to hyperscale data center platforms. It reports through distinct businesses: Client Computing Group (CCG), Data Center & AI (DCAI), Intel Foundry Services (IFS), and corporate/other. For investors the DCAI segment is the growth engine: custom silicon and AI accelerators for cloud providers and enterprises are the clearest path to sustainable margin upside. Foundry is strategically important because it turns Intel into a multi-customer manufacturer - but it's currently loss-making and capital intensive.
The numbers that matter
- Q2 revenue: $16.1 billion (reported), up 25% year-over-year.
- DCAI growth: +59% year-over-year, the fastest growth vector in the quarter.
- IFS (foundry) revenue: $5.8 billion, up 31% year-over-year, still reporting a $2.1 billion operating loss.
- Adjusted EPS: $0.42, roughly double street expectations according to reported commentary.
- GAAP net loss: $11.03 billion, driven by a $12.5 billion non-cash CHIPS Act charge that masked operating progress.
- Free cash flow (TTM or reported): $2.83 billion; capex guidance for 2026: $20 billion (management expects 2027 capex to be larger).
- Market capitalization: roughly $411 billion and enterprise value around $450.7 billion.
- Valuation metrics: price-to-sales ~7.2, EV/EBITDA ~27.0, price-to-book ~4.7. Trailing reported EPS is negative, reflecting accounting volatility.
- Share price context: current $85.24; 52-week high $142.35, low $18.97. Technicals show RSI ~31.5 (near oversold) and bearish MACD momentum.
Why this matters now
Two structural dynamics are colliding: hyperscale AI demand for servers (favoring Intel's x86 compatibility and growing custom AI processors) and Intel's strategic pivot to be a competitive foundry. The DCAI growth of 59% shows actual commercial demand for Intel's cloud and AI silicon. Meanwhile, named foundry customers are starting to appear - Fortinet was announced as a first named foundry win - which validates the go-to-market effort even as the unit digests heavy investment and operating losses. If DCAI keeps compounding and IFS can convert more customers while reducing losses, the operating leverage could be meaningful.
Valuation framing
On headline multiples Intel looks expensive relative to historical troughs because market cap sits at $411 billion while free cash flow remained modest at $2.83 billion. EV/EBITDA of ~27 and price-to-sales >7 reflect a market that has already priced a significant amount of AI growth into the shares. But a few caveats soften that picture: the GAAP loss is driven largely by a non-cash charge, adjusted EPS was positive, and DCAI growth is real. This is a stock where valuation is concentrated in future margin expansion assumptions rather than current cash conversion.
Consider this: if Intel's foundry business cures its heavy losses and DCAI margins expand in line with peers, multiple expansion is plausible. If not, the large capex profile and thin free cash flow make the current level vulnerable. That duality is why a trade with defined entry/stop/target makes sense.
Actionable trade plan
| Plan item | Detail |
|---|---|
| Entry price | $85.24 |
| Stop loss | $75.00 |
| Target | $110.00 |
| Trade direction | Long |
| Horizon | Long term (180 trading days) - hold into the next several earnings cycles to see whether DCAI growth sustains and foundry losses narrow. |
Rationale for horizon: Intel needs multiple quarters of consistent revenue and margin improvement for the market to re-rate the shares. The long-term (180 trading days) horizon gives the company time to convert customer wins, show progress on foundry cost reductions, and prove the durability of AI-driven server demand.
Catalysts to watch (2-5)
- Quarterly DCAI revenue prints and sequential margin improvement - continued double-digit growth would validate the thesis.
- New named foundry customers and multi-year supply agreements that demonstrate commercial traction beyond Fortinet.
- Management guidance on capex cadence and evidence that incremental investment is being absorbed into profitable capacity rather than dilutive spending.
- Contract wins or design wins for Intel custom AI accelerators at hyperscalers that increase addressable revenue per server.
Risks and counterarguments
This trade is not without significant downside risk. Key risks include:
- Capex shock and cash consumption - management guided to $20 billion of capex for 2026 and said 2027 will be significantly higher. If capex outpaces cash generation, equity could face dilution or multiple compression.
- Foundry execution risk - IFS is still unprofitable (reported $2.1 billion loss). If process yields, cost curves, or customer uptake lag, the business could remain a cash drain longer than anticipated.
- Macro demand variability - AI server cycles are front-ended by hyperscalers; any slowdown in enterprise or cloud spending would cut DCAI growth sharply.
- Valuation already reflects some upside - multiples (EV/EBITDA ~27, price-to-sales ~7.2) imply the market expects sizable margin recovery; disappointment could produce sharp downside.
- Headline volatility from non-cash accounting items - large one-time charges (e.g., CHIPS Act accounting) can produce headline losses even while underlying operations improve, driving kneejerk selling.
Counterargument: You could argue Intel is already priced for perfection in AI because recent strength and the Fortinet foundry win are in the price. Multiples are not cheap relative to the company's current free cash flow, and sustained capex and potential dilution create asymmetric downside. That view is defensible and would make me favor smaller position sizing or waiting for clearer margin expansion before adding exposure.
How I'll know I'm wrong
This trade will be invalidated if any of the following occur over the next 180 trading days:
- Quarterly DCAI growth falls materially below current momentum (e.g., single-digit growth or sequential decline), implying demand was episodic rather than structural.
- Foundry losses widen or yield problems persist, demonstrating the unit will remain a multi-year cash drain without a path to profitability.
- Management raises capex again without clear guidance on how that spending will convert to higher margins or multi-year contracts.
Position sizing and trade discipline
This setup suits investors who are comfortable with a medium-sized allocation within a diversified portfolio. Given the combination of operational improvement and headline capex risk, consider risking no more than 1-3% of portfolio capital on this single trade and use the $75 stop as a hard exit to preserve capital if the thesis fails.
Conclusion
Intel's operational results show the turnaround is real: DCAI is growing fast, foundry is starting to win named customers, and adjusted EPS was positive. The market's negative reaction to the quarter reflects legitimate concerns about capex and the near-term earnings hit from non-cash accounting items. That pessimism creates a tactical opportunity: buy now at $85.24 with a $75 stop and a $110 target over the next 180 trading days if you believe AI infrastructure demand remains durable and Intel can bend its foundry losses toward breakeven.
If the company proves foundry economics and shows sustained DCAI margin expansion, multiple expansion to justify the target is a reasonable outcome. If it fails to do that, the stop is appropriately placed to limit downside while the long-term secular AI cycle plays out elsewhere in the portfolio.