Looking for a Good Deal on an AI Stock? This Leading Chipmaker Is Cheaper Than Broadcom and AMD.

Nvidia is presented as an attractive AI investment, trading at 27x trailing earnings versus 131x for AMD, 46x for Broadcom, and roughly 33x for the technology sector. Fiscal Q2 revenue more than doubled year over year to $96.2 billion and net income rose 126% to $59.7 billion, while Q3 revenue guidance of $108 billion implies 90% annual growth. The article argues Nvidia's estimated 86% GPU market share and continued hyperscaler AI infrastructure spending support its growth outlook despite rising competition from Broadcom's custom AI chips and AMD's data-center offerings.

Analysis

The relevant valuation question is not trailing P/E but the durability of accelerator gross profit after the current build cycle. NVDA's premium ecosystem economics—software lock-in, networking attach and rapid platform cadence—can sustain a higher multiple than a merchant-chip peer, but its earnings base is now unusually exposed to a small set of buyers maintaining capex intensity. A deceleration in hyperscaler capex growth, even with spending still rising in absolute dollars, would likely compress NVDA's multiple before it materially affects reported revenue; this is the principal 1-3 month risk around quarterly guidance.

AVGO is the cleaner diversification hedge within AI silicon. Custom ASIC adoption does not require customers to abandon GPUs; it is most economical for stable, high-volume inference workloads, while frontier-model training and fast-changing architectures favor programmable NVIDIA systems. The second-order beneficiary is AVGO's networking franchise, since scale-out custom silicon still raises demand for high-bandwidth switching and interconnect; the risk is that bespoke programs have concentrated customers, long design cycles and weaker optionality if a major program slips.

AMD's upside requires share gains and software/customer-validation progress rather than simply a larger AI market. The claimed CPU intensity increase from agentic workloads is directionally supportive, but it also makes platform-level competition from ARM-based servers and hyperscaler-designed silicon more relevant. Consensus is likely underpricing the distinction between AI infrastructure spend and investable semiconductor profit pools: inference can broaden total silicon demand while simultaneously shifting mix away from the highest-margin GPU configurations over the next 6-18 months.

The article's relative-value framing is too simplistic: NVDA's lower trailing P/E may reflect exceptional current earnings rather than a discount, while AVGO's reported P/E is affected by acquisition accounting and its non-AI businesses. Treat any retail-driven reaction as low-information absent corroboration from hyperscaler capex guidance, NVDA backlog conversion, and evidence that ASIC deployments are incremental rather than substitutive.

Key Decisions for Investors

• Maintain/enter a core long NVDA only on post-results weakness rather than chasing a positive commentary catalyst; use a 3-6 month horizon. Favor a defined-risk call spread over outright calls if implied volatility is elevated. Thesis fails if next-quarter revenue guidance or gross-margin trajectory signals a material platform-transition or supply-demand digestion.
• Initiate a 6-12 month pair: long AVGO / short AMD in equal dollar beta-adjusted exposure. AVGO has a more credible pathway to monetize both custom compute and networking, whereas AMD needs sustained accelerator share gains to justify higher expectations. Reassess if AMD demonstrates consecutive quarters of material data-center margin expansion and named large-scale accelerator wins.
• Use GOOG and META capex guidance as the highest-frequency read-through for the group: reduce NVDA/AVGO gross exposure if either signals lower 2027 infrastructure growth or emphasizes internally designed silicon displacing external procurement. This is an alert condition, not a standalone short signal, because capacity commitments can keep purchase orders elevated temporarily.
• For a more defensive AI expression over 6-18 months, prefer a basket long AVGO and networking exposure versus a short on GPU-pure-play beta only after evidence that inference workloads are scaling faster than training. The key confirmation is rising ASIC/networking revenue alongside flattening accelerator unit growth; absent that data, avoid assuming custom silicon is immediately cannibalistic to NVDA.