1 Major Reason Why Broadcom Still Has More Room to Run

Broadcom's AI semiconductor revenue reached $8.4 billion in fiscal Q1, up 106% year over year, with CEO Hock Tan saying the custom AI chip business could exceed $100 billion in annual revenue by next year. Wall Street is projecting total Broadcom revenue of $158 billion by fiscal 2027, up from $64 billion in 2025, supporting a bullish long-term valuation case. The article is opinionated rather than event-driven, but it reinforces positive sentiment around Broadcom's AI growth trajectory.

Analysis

AVGO is increasingly functioning like a picks-and-shovels monopolist on AI inference economics, not just an HPC supplier. If custom silicon keeps taking share, the second-order winner is the platform owner with the deepest deployment footprint and software lock-in: hyperscalers that can amortize design costs across massive traffic and use lower-cost silicon to defend margins. That is mildly negative for GPU attach rates at the margin, but not catastrophic for NVDA unless custom chips broaden from tightly constrained workloads into a much larger share of training spend.

The market is likely underestimating the operating leverage embedded in Broadcom’s AI mix. The key issue is not whether AI revenue grows, but whether supply and customer concentration allow AVGO to convert that growth into sustained free cash flow without overpromising on timing; any slippage in tape-out, packaging, or customer ramps would matter more than the headline growth rate. A multi-quarter cadence of beat-and-raise is likely the real catalyst, because the stock’s multiple can expand further only if investors believe the revenue bridge is executable, not aspirational.

The contrarian read is that the bull case is already crowded and increasingly dependent on a very narrow set of hyperscaler capex decisions. If one or two customers pause deployments for inventory digestion or internal yield issues, the downside can be sharp because consensus will have implicitly capitalized near-perfect execution over the next 4-6 quarters. GOOGL is the cleaner second-order beneficiary because lower-cost custom silicon supports AI pricing power and cloud margin defense; NVDA remains the strategic loser only if custom workloads prove sticky enough to move beyond niche optimization and into mainstream inference.

Near term, this is a sentiment trade as much as a fundamentals trade: the stock can keep grinding higher over weeks, but the asymmetry improves on any pullback tied to macro or semiconductor de-risking. The main tail risk is that expectations outrun the actual revenue conversion path, creating a classic high-multiple/mid-cycle disappointment even while fundamentals remain strong.

Key Decisions for Investors

• Stay long AVGO, but size for momentum rather than convexity: add on 3-5% pullbacks over the next 2-6 weeks, with a 6-12 month target tied to continued AI revenue beats; trim into sharp gap-ups as the market may be front-running the 2027 ramp.
• Pair trade: long AVGO / short NVDA into any broad AI semis rally over the next 1-3 months, on the thesis that custom silicon share gains are a relative headwind to GPU multiple expansion even if NVDA fundamentals stay solid.
• Express the second-order winner via GOOGL over a 3-9 month horizon: long GOOGL calls or stock against a weaker AI-infrastructure basket, since lower custom-chip costs should support cloud economics and AI monetization without requiring heroic growth assumptions.
• Avoid chasing AVGO after outsized gap moves; instead sell cash-secured puts 10-15% below spot with 60-90 day tenor to monetize elevated sentiment while keeping entry discipline if execution hiccups create a drawdown.
• If AI semis sell off on capex fears, use that as a tactical add signal for AVGO rather than a de-risk point; the business model is more resilient than the stock’s day-to-day beta implies.