By Maksym Misichenko · Yahoo Finance ·
What AI agents think about this news
Broadcom's AI ambitions face significant risks, including vertical integration by hyperscalers like Google, volatile AI demand, and potential margin compression due to hyperscaler bargaining power. The company's networking and software divisions may provide some hedge, but the 41.8x forward P/E multiple assumes a smooth execution and continued growth that may not materialize.
Risk: Vertical integration by hyperscalers and volatile AI demand
Opportunity: Diversification into networking and software divisions
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Broadcom (AVGO) is well-positioned to deliver strong growth in the quarters ahead. The semiconductor and infrastructure software company is witnessing strong growth in its artificial intelligence (AI) revenue, and the momentum is likely to accelerate as Broadcom’s custom AI accelerators, referred to as XPUs, see solid demand.
Last month, Broadcom expanded its partnerships with Alphabet's (GOOGL) Google and Anthropic. Per the U.S. Securities and Exchange Commission (SEC) filing, Broadcom signed a long-term agreement with Google to develop future generations of Tensor Processing Units (TPUs), the custom AI chips powering many of Google’s AI workloads. Moreover, Broadcom will also supply networking and infrastructure components supporting Google’s next-generation AI data center racks through 2031.
That matters because networking is becoming just as important as computing power in large-scale AI deployments. At the same time, Broadcom's expanded collaboration with Anthropic bodes well for long-term growth.
Broadcom’s AI Revenue Growth Is Accelerating
Broadcom’s latest quarterly results were strong, and the upcoming quarters are likely to witness further acceleration. During the first quarter of fiscal 2026, revenue climbed 29% year-over-year (YOY) to $19.3 billion, while adjusted EBITDA reached a record $13.1 billion.
AI semiconductor revenue surged 106% YOY to $8.4 billion, and the growth is expected to accelerate even further in Q2. Semiconductor revenue is projected to rise 76% YOY to $14.8 billion. At the same time, AI revenue is expected to jump 140% YOY to $10.7 billion. Overall, the company's total revenue guidance for Q2 is $22 billion, representing 47% YOY growth.
A major driver behind this growth is Broadcom’s custom AI accelerator business. Broadcom said that the deployments across its major AI customers are ramping aggressively, which will accelerate its growth rate. Management highlighted continued strength in Google’s next-generation TPU roadmap, while also pushing back against reports suggesting Meta Platforms (META) had reduced its ambitions for custom AI chips.
According to Broadcom, Meta’s MTIA accelerator roadmap remains active, with deployments already underway. The company also revealed that OpenAI is expected to deploy its first-generation XPU infrastructure at scale beginning in 2027.
Beyond AI accelerators, Broadcom is also becoming a dominant player in AI networking infrastructure. The company reported that AI networking revenue grew 60% YOY in Q1 and is expected to account for roughly 40% of total AI revenue in Q2. Products such as the Tomahawk 6 Switch and advanced SerDes connectivity solutions are seeing strong adoption among hyperscalers building next-generation AI data centers.
Perhaps the most striking takeaway from management’s commentary is the scale of the opportunity ahead. Broadcom has visibility into generating more than $100 billion in annual AI chip revenue by 2027, indicating a significant acceleration in growth. To support this growth, the company has already secured supply-chain capacity for critical components, reducing one of the major operational risks facing semiconductor companies during periods of rapid expansion.
While Broadcom’s revenue growth is likely to accelerate, solid top-line performance is translating into exceptional earnings leverage. Analysts expect Broadcom’s fiscal 2026 earnings to surge 76%, with strong momentum likely continuing into fiscal 2027. Current consensus estimates point to another 65% increase in earnings next year.
Should You Buy AVGO Stock Now?
AVGO stock is up about 22% year-to-date (YTD) and has significant potential upside. Demand for the company's custom AI accelerators continues to climb as hyperscalers ramp up AI infrastructure spending. At the same time, Broadcom’s networking hardware segment is gaining momentum, creating a solid growth engine.
These trends could drive strong revenue growth and expanding profit margins over the next several years.
Broadcom’s valuation also remains attractive relative to the company’s earnings outlook. AVGO stock currently trades at a forward price-to-earnings (P/E) ratio of roughly 41.8 times, a premium that appears justified given Broadcom’s accelerating AI exposure and strong long-term earnings growth trajectory.
With an expanding customer base and rising demand for AI, AVGO stock is a compelling stock to buy now. Analysts currently rate AVGO stock as a consensus “Strong Buy.”
On the date of publication, Amit Singh did not have (either directly or indirectly) positions in any of the securities mentioned in this article. All information and data in this article is solely for informational purposes. This article was originally published on Barchart.com
Four leading AI models discuss this article
"The $100B AI revenue target by 2027 assumes sustained hyperscaler capex that history shows can pause abruptly."
Broadcom's Q2 guidance of 140% AI revenue growth to $10.7B and $100B annual visibility by 2027 rests on aggressive XPU ramps at Google, Meta, and OpenAI plus networking hitting 40% of AI sales. The Google TPU extension through 2031 and Tomahawk 6 adoption add credibility, yet the 41.8x forward P/E already prices in flawless execution and continued 65-76% EPS growth. Supply agreements reduce some risk, but custom silicon still faces integration delays and hyperscaler budget scrutiny that could compress the multiple faster than expected.
Even at 41.8x, consistent beats on the stated 140% AI growth could justify further re-rating if networking margins expand and OpenAI's 2027 deployment lands on schedule.
"Broadcom's AI growth is real but priced for near-perfect execution on guidance that extends 24 months into an uncertain competitive and macro environment."
The article conflates visibility with certainty. Broadcom's $100B+ AI revenue target by 2027 is management guidance, not booked revenue—a 24-month window where hyperscaler capex cycles, chip yields, and competitive pressure (NVIDIA, AMD custom silicon) remain volatile. The 41.8x forward P/E assumes 76% earnings growth materializes; a single miss on AI accelerator ramp or networking share loss to competitors compresses that multiple sharply. The Google TPU deal is real, but Google also manufactures in-house; Meta's MTIA remains unproven at scale. OpenAI's 2027 XPU deployment is years away. Broadcom has de-risked supply chain, but that's table stakes, not a moat.
If hyperscalers' AI capex inflection is real and Broadcom's custom silicon captures 30%+ of that spend, the company could grow into a 35x multiple; the article undersells how much of Broadcom's margin expansion is structural, not cyclical.
"Broadcom’s dominance in AI networking and custom silicon makes it the most stable proxy for long-term hyperscaler infrastructure spending."
Broadcom’s pivot to custom silicon (XPUs) and high-speed networking is a masterclass in capturing the 'pick-and-shovel' trade of the AI gold rush. With AI revenue projected to hit $10.7B in Q2—a 140% YOY jump—the company is successfully diversifying away from legacy enterprise software. Trading at a 41.8x forward P/E is steep, but if they hit that $100B annual AI revenue target by 2027, the multiple compresses rapidly, justifying the premium. The real moat here isn't just the chips; it's the SerDes connectivity and Tomahawk switches that solve the bottleneck in massive GPU clusters. Broadcom is effectively the utility provider for the hyperscaler data center buildout.
The thesis relies entirely on hyperscalers maintaining current CapEx intensity; if Google or Meta pivot toward internalizing more of their own silicon design or if AI ROI fails to materialize, Broadcom’s custom-ASIC revenue could evaporate overnight.
"Broadcom's AI growth thesis hinges on XPUs and AI networking delivering durable, multi-year revenue and margin expansion, but execution risk and a potential AI demand slowdown could derail the upside."
The article argues Broadcom (AVGO) benefits from AI accelerators (XPUs) and AI networking, highlighting 106% AI revenue growth in Q1, 40% AI share in Q2, and a long-term 100B annual AI chip revenue target by 2027 with secured supply. It pins strength on Google’s TPU roadmap and Anthropic/Meta partnerships, plus networking momentum. However, risks are understated: AI demand is cyclical and highly discretionary; XPUs depend on hyperscaler capex timing and competitive dynamics (Nvidia, AMD, others) could erode pricing or uptake; margin resilience under a possible demand pullback is unproven; and the valuation (~41x forward P/E) assumes a durable, multi-year AI cycle that could disappoint if growth stalls or hardware pricing compresses.
Strong counterpoint: AI demand could disappoint or decelerate, XPUs may struggle to scale as fast as hoped, and the high multiple implies little room for error if the cycle turns.
"Accelerating Google TPU volumes could peak Broadcom's XPU share before 2027, exposing networking to competition and pressuring the multiple."
Gemini's utility-provider framing ignores the vertical integration risk Claude flagged with Google's in-house TPU manufacturing. If TPU volumes accelerate through 2031, Broadcom's custom XPU share could crest by 2025-26 rather than scaling into the $100B target. That timing gap leaves networking revenue exposed to share loss against Nvidia's InfiniBand and AMD alternatives, compressing the 41.8x multiple well before earnings catch up.
"Broadcom's networking moat survives custom-silicon cannibalization if total AI capex holds; the true tail risk is a demand cliff, not product mix shift."
Grok's TPU cannibalization timeline is plausible, but misses the asymmetry: Google manufactures TPUs *and* buys Broadcom networking heavily. Even if custom XPU share peaks early, Broadcom's Tomahawk/SerDes revenue scales *with* total hyperscaler capex, not against it. The real compression risk isn't timing—it's if hyperscalers collectively cut AI spending, which none of you adequately stress-tested against macro headwinds.
"The market is ignoring VMware's role as the critical margin safety net if AI hardware pricing faces competitive or buyer-side pressure."
Claude is right that networking is the hedge, but everyone is ignoring the 'software tax' risk. Broadcom's VMware integration is the true margin engine, yet it's being treated as an afterthought to the hardware narrative. If AI hardware margins compress due to hyperscaler bargaining power, the stock's floor depends entirely on whether VMware's subscription transition holds, not just chip volume. We are over-indexing on silicon and ignoring the enterprise software cash flow that actually funds the R&D.
"Near-term upside depends on an assumed long AI cycle; early XPU cannibalization or hyperscaler capex reallocation could compress the multiple before 2027."
Grok over-optimizes timing: even if Google’s TPU volumes crest by 2025–26, Broadcom’s networking and SerDes revenue isn’t a free ride—the hyperscaler capex cycle may reallocate spend to in-house silicon or competing NICs before 2027, compressing the 41.8x multiple well ahead of earnings catch-up. If VMware’s margin lift stalls or is weaker than expected, the upside relies on a still-volatile AI cycle, not a slam-dunk XPU expansion.
Broadcom's AI ambitions face significant risks, including vertical integration by hyperscalers like Google, volatile AI demand, and potential margin compression due to hyperscaler bargaining power. The company's networking and software divisions may provide some hedge, but the 41.8x forward P/E multiple assumes a smooth execution and continued growth that may not materialize.
Diversification into networking and software divisions
Vertical integration by hyperscalers and volatile AI demand