Should You Buy Intel Stock After a Nearly Fourfold Year?

Intel's Q2 revenue rose 25% year over year to $16.1 billion, while Data Center and AI operating income nearly quadrupled to $2.5 billion on 59% revenue growth and roughly 40% operating margins. However, Intel Foundry still posted a $2.1 billion operating loss, with only $293 million of external revenue, and the company issued 242 million shares for about $23 billion, lifting its share count roughly 20% year over year. At about $96 per share, Intel trades near 47x expected 2027 earnings versus roughly 20x for TSMC, leaving limited valuation support unless AI growth, foundry breakeven, and external customer wins all materialize.

Analysis

INTC’s rerating now embeds three independent underwriting assumptions: sustained server-CPU share stabilization, data-center margin durability, and a credible path to third-party leading-edge foundry utilization. The key vulnerability is that these are not fully additive—higher product volumes consume internal wafer capacity while external foundry credibility requires customer commitments, yield transparency, and economics that can stand without captive demand. Incremental capex before external-volume proof raises the risk that operating improvement converts poorly into per-share free cash flow, particularly after the enlarged share base.

The relevant competitive comparison is not merely TSM’s lower earnings multiple; it is the valuation of execution certainty. TSM’s advanced-node ecosystem, packaging capacity, and customer diversification provide a cleaner way to own AI semiconductor capex, while INTC remains exposed to a CPU demand cycle and a manufacturing turnaround simultaneously. FTNT’s design win has limited read-through for Intel’s strategic node roadmap: an older-node security ASIC does not validate leading-edge yields, PDK maturity, or high-volume customer willingness to dual-source away from TSM.

Over the next 1-3 months, the decisive catalyst is whether management can identify meaningful external 14A commitments and quantify associated wafer volumes, prepayments, or capacity reservations. Over 6-18 months, monitor gross-margin progression against capex and whether data-center gains reflect unit share rather than a cyclical server refresh; NVDA’s platform pull and AMD’s EPYC roadmap can limit Intel’s ability to retain improved CPU pricing. The contrarian upside is that a government-backed balance sheet and a real 14A anchor customer could compress the execution discount rapidly, making an outright short vulnerable to sharp squeezes.

Key Decisions for Investors

• Initiate a 3-6 month pair: long TSM / short INTC in equal dollar amounts. The thesis is multiple convergence through execution-quality differentiation, not a broad semiconductor call; target 15-25% relative performance. Cover if INTC announces a credible high-volume 14A customer with disclosed capacity reservation or if TSM faces a material Taiwan-risk escalation.
• For directional exposure, buy INTC 6-month put spreads rather than naked short stock, centered around 10-15% below spot. This limits squeeze risk while positioning for a valuation reset if third-quarter guidance or foundry-loss commentary shows that capex is outrunning monetization; exit if forward EPS consensus rises materially without another increase in share count.
• Maintain or add TSM on weakness ahead of foundry/customer updates, but size geopolitical tail risk explicitly. A durable reduction in TSM’s advanced-node pricing, utilization, or CoWoS demand would falsify the relative-quality thesis more clearly than headline comparisons of P/E multiples.
• Do not treat FTNT as a meaningful Intel foundry beneficiary absent disclosed production volumes and node economics. Use any market extrapolation from the design win into FTNT as a watch item rather than a trade signal.