Broadcom Could Be The Biggest AI Winner Nobody Is Talking About

Broadcom reported Q2 fiscal 2026 revenue of $22.19 billion, up 48% year over year and above estimates by $70 million, while adjusted EPS rose 54% to $2.44 and beat consensus by $0.04. The stock fell 15% after management's Q3 AI chip revenue guide of $16 billion missed the $17.2 billion consensus, though total Q3 revenue guidance of $29.4 billion still implies 83% growth. Management reiterated a $100 billion AI chip revenue target for fiscal 2027, keeping the long-term AI thesis intact despite near-term softness.

Analysis

The market is treating this as a guidance miss, but the more important signal is that the AI buildout is shifting from “any growth is good growth” to a curve-fitting problem where buyers will punish deceleration even at very high absolute levels. That usually marks a transition from multiple expansion to multiple compression in the hottest supplier names, even when the long-duration thesis is intact. For AVGO, the issue is not demand disappearing; it is that the stock was priced for near-perfect sequencing, so a small wobble in one quarter can reset expectations for several quarters.

Second-order, this is more about capex mix than AI demand. If hyperscalers are still spending but reallocating toward custom silicon, the beneficiary set broadens from GPU vendors to design/IP and networking-adjacent names, while pure-play accelerator leaders can see pressure in narrative even if unit economics remain strong. That creates a subtle read-through: a slower custom-chip ramp can temporarily help NVDA sentiment on the margin by reducing fears of immediate share loss, while META and GOOGL remain structurally advantaged if they can keep migrating workloads to lower-cost silicon.

The contrarian angle is that the selloff may be overdone on a one-quarter sequencing miss because the revenue base is now so large that even a modest miss in the AI line still implies extreme growth. The real risk is not fundamentals over the next 12 months, but whether investors start demanding proof that the 2027/2028 model is financeable without a second wave of hyperscaler capex acceleration. If that proof slips, the stock can stay range-bound for months even if earnings keep compounding.

Catalyst-wise, the next 1-2 quarters matter more than the annual target: if management can show order visibility and mix improvement, the post-earnings de-rating should partially reverse. If not, the market will start comparing AVGO to other “AI infrastructure” names on forward growth momentum rather than long-term TAM, which historically compresses multiples before the fundamentals crack. The setup favors tactical patience rather than chasing the first dip.

Key Decisions for Investors

• Trim AVGO on strength over the next 1-2 weeks; if it stabilizes after the post-earnings de-risking, re-enter only on evidence of a second-quarter rebound in AI bookings. Risk/reward is now better for waiting than for averaging down immediately.
• Use a pair trade: long GOOGL or META vs short AVGO for 1-3 months. If hyperscaler capex remains firm, the market should prefer the end-customer/beneficiary layer over the supplier exposed to sequencing risk; target a 10-15% relative move with defined stop if AVGO reaccelerates.
• Consider a short-dated AVGO call spread only if the stock base-builds for several sessions after the selloff. The skew is likely still rich, so upside participation is cheaper via spreads than outright stock if the thesis is a normalization rally, not a straight-line recovery.
• Avoid chasing NVDA weakness on this headline alone; the read-through is mixed and could actually support NVDA sentiment if investors conclude custom ASIC adoption is still incremental rather than substitutional. Maintain NVDA as a relative long versus AVGO on a 1-2 month horizon.
• For longer-term portfolios, accumulate AVGO only on a further 10-15% de-rating or on confirmation that the next guide-up cycle is intact. At current levels, the stock still has upside, but the margin of safety is thinner than the market narrative suggests.