Prediction: This Artificial Intelligence (AI) Chip Stock Will Make a Big Move in October (Hint: It’s Not Micron)

Marvell Technology is expected to raise guidance at its Oct. 6 Investor Day, with RBC Capital Markets forecasting at least a $2 billion increase to its fiscal 2029 AI-revenue outlook from the current $10 billion estimate. Marvell has already lifted fiscal 2027 revenue guidance to $12 billion from $11.5 billion and fiscal 2028 guidance to $18 billion, while long-run EPS could reach $15-$20 if operating margins expand. Despite trading at 86x trailing and 62x forward earnings, the article argues its 16% pullback from the June 18 high presents an opportunity, citing potential for the shares to nearly double by 2030.

Analysis

MRVL’s setup is increasingly a credibility-and-execution trade rather than a pure AI-beta trade. A further long-range target increase could support the shares for days, but the valuation already capitalizes a large portion of the proposed earnings ramp; the relevant question is whether management provides nearer-term revenue conversion, customer concentration, design-win timing, and gross-margin evidence. Without those disclosures, a larger TAM figure is unlikely to sustain multiple expansion beyond the event.

The more differentiated exposure is in optical interconnect and custom silicon, where MRVL competes for a larger share of AI infrastructure spend that otherwise accrues to NVDA, AVGO and networking vendors. A broadening AI capex cycle would also reinforce MU demand, but MRVL’s margins are more exposed to mix, yield and customer-specific product ramps; a delay in one hyperscaler program can matter disproportionately. This makes MRVL higher-beta than MU but also more vulnerable if hyperscalers prioritize proven merchant platforms or defer network upgrades.

Consensus appears to be extrapolating management’s long-term framework into a smooth earnings trajectory. The likely underappreciated risk is that successive target raises pull forward valuation while leaving fiscal-year estimates insufficiently revised; in that case, even a bullish Investor Day can produce a sell-the-news reaction. Over the next 1-3 months, estimate revisions and disclosed backlog/design-win milestones matter more than TAM rhetoric; over 6-18 months, optical attach rates and custom-ASIC gross-margin progression determine whether MRVL can retain a premium multiple.

Key Decisions for Investors

• Event trade only if options imply a move below the stock’s recent post-earnings/event range: buy a defined-risk MRVL call spread expiring 4-8 weeks after Oct. 6 rather than outright shares. Take profits on a guidance-driven gap if FY27/FY28 consensus revisions do not follow within two weeks; maximum loss is premium paid.
• For fundamental exposure, prefer a 1-3 month pair of long MRVL / short SOXX or SMH, sized beta-neutral. This isolates the company-specific upside from validated custom-silicon and optical-networking milestones while reducing broad AI-semiconductor multiple risk; exit if management does not quantify incremental revenue timing or if forward revenue estimates fail to rise.
• Use MU as the cleaner AI-memory confirmation vehicle rather than treating its strength as proof of MRVL execution. Maintain or initiate MU exposure only while data-center pricing and bit-demand commentary continue to improve; a weakening memory-price outlook would be an early warning that infrastructure spend is becoming more selective.
• Do not underwrite long-term EPS or valuation targets until MRVL discloses customer concentration, production-ramp cadence, and segment margin assumptions. A failure to provide those items, or a meaningful reduction in gross-margin trajectory at the next earnings update, falsifies the premium-multiple thesis and warrants reducing MRVL exposure.