By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
Panelists agree that Broadcom's strong AI chip and networking growth, high gross margins, and Google TPU deal provide near-term momentum, but the stock's high valuation (81x trailing P/E) may not be sustainable given potential risks such as customer concentration, margin compression, and a slowdown in hyperscaler capex.
Risk: Margin compression due to customer negotiations and a slowdown in hyperscaler capex, as well as customer concentration risk.
Opportunity: None explicitly stated.
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 is winning in two ways at once -- custom AI chip sales surged 140% in Q1, while data-center networking revenue jumped 60%.
Unlike cash-hungry AI peers, Broadcom has 77% gross margins, which have fueled a 15-year dividend growth streak and a new $10 billion buyback.
Broadcom shares are expensive, but its AI tailwinds and a new long-term deal with Google help to justify the premium.
There are a lot of great AI stocks to choose from these days, but they don't all offer the same long-term opportunities or tap into artificial intelligence (AI) trends as Broadcom (NASDAQ: AVGO). Nor do they pay a growing dividend, as the company also does.
Broadcom is benefiting from two big AI trends, including a surge in data center demand and shift by big tech companies toward customized processors.
Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »
And it's likely that Broadcom's impressive run isn't done yet. Here's why.
AI is a vast market, so let's narrow it down to exactly how Broadcom benefits from it. First, its custom AI processors are increasingly in demand by leading tech companies to optimize their AI systems.
Broadcom's processors can be designed specifically with a tech company's large language models in mind, making them more efficient and boosting performance. This advantage helped its custom AI application-specific integrated circuits (ASICs) have a 140% increase in sales in the first quarter of 2026, thanks to expanding deals with Alphabet, Meta, Anthropic, and OpenAI.
The chipmaker recently announced that it has expanded its partnership with Alphabet's Google, which keeps Broadcom designing Google's tensor processing units (TPUs) and supplying networking components through 2031.
Broadcom's AI-specific semiconductor revenue more than doubled in the first quarter to $8.4 billion, accounting for about 43% of its total revenue. And more good times are on the way with management estimating its AI chip revenue will reach $100 billion by 2027.
And that's not the only AI trend benefiting the business. The company's networking systems -- mainly through its popular Tomahawk switches for data centers -- are also experiencing a surge in demand as hundreds of billions of dollars are invested in data center expansion.
Broadcom's AI networking revenue rose 60% in the most recent quarter to $2.8 billion -- accounting for about one-third of all the company's AI sales. And management believes this will continue to expand, with networking sales making up 40% of AI revenue in the second quarter.
High-growth AI stocks aren't typically the companies you would expect to pay dividends, which is one of the ways Broadcom is unique.
The company returned $3.1 billion to shareholders in the first quarter -- or $0.65 per share -- and has a history of raising its dividend. It has increased it annually for 15 consecutive years. Management also repurchased $7.8 billion in shares in the quarter and authorized an additional $10 billion repurchase program through 2026.
While many AI companies are focused on reinvesting earnings in research and development (R&D) rather than paying dividends, Broadcom's high profitability and soaring revenue allow it to do both.
The company ended the first quarter with $14.2 billion in cash and cash equivalents and had consolidated gross margins of 77%. That cash and high profitability make it easy for Broadcom to return cash to shareholders, while still investing $1.5 billion into R&D in the quarter.
It will probably come as no surprise that Broadcom's stock isn't exactly cheap right now. Most AI stocks aren't.
After its shares climbed 79% over the past year, the trailing price-to-earnings ratio (P/E) of nearly 81 is significantly higher than the tech sector's average P/E of 36.
Still, with Broadcom clearly tapping into established -- and very profitable -- AI trends and securing new deals that should help keep its current momentum going, investors probably won't regret adding some Broadcom stock to their portfolio. And the company's growing dividend is just icing on the cake.
Before you buy stock in Broadcom, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and Broadcom wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $477,813! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,320,088!
Now, it’s worth noting Stock Advisor’s total average return is 986% — a market-crushing outperformance compared to 208% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
**Stock Advisor returns as of May 23, 2026. *
Chris Neiger has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Alphabet, Broadcom, and Meta Platforms. The Motley Fool has a disclosure policy.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.
Four leading AI models discuss this article
"Broadcom's AI momentum is real, but the 81x P/E already embeds flawless execution through 2027."
Broadcom's 140% custom ASIC surge and 60% networking growth in Q1 underscore real AI demand, with 77% gross margins enabling both $3.1B quarterly returns and $1.5B R&D spend. The Google TPU extension to 2031 and $100B AI revenue target by 2027 add visibility. Yet the 81x trailing P/E after a 79% run already prices in aggressive execution, leaving little margin if hyperscaler capex slows or custom-chip competition from Marvell or in-house designs intensifies. Concentration in four clients and networking's projected rise to 40% of AI sales introduce single-point risks the article downplays.
The $100B target and 15-year dividend streak could still drive multiple expansion if Q2 networking beats confirm the 40% mix shift, making today's premium look cheap in hindsight.
"Broadcom's AI tailwinds are real, but the stock prices in $100B AI revenue by 2027 with zero margin of safety for customer concentration or in-house chip development."
Broadcom's 140% custom AI chip growth and 77% gross margins are genuinely impressive, and the Google TPU deal through 2031 provides real revenue visibility. However, the article conflates two separate narratives: near-term momentum (Q1 2026 results) with long-term valuation. At 81x trailing P/E against a 36x tech average, you're pricing in flawless execution through 2027 when AI chip revenue allegedly hits $100B. That's a 12x revenue multiple on an unproven segment. The dividend is window dressing—$3.1B returned quarterly on what gross profit base? The real risk: if any major customer (Google, Meta) brings AI chip design in-house or diversifies suppliers, Broadcom's concentration risk explodes. The article doesn't quantify customer concentration.
If Google, Meta, and others have already committed to custom ASICs and Broadcom owns the networking layer too, the 81x multiple might compress toward 50-60x as growth normalizes—still 40% downside from here, even if the thesis holds.
"Broadcom's current valuation leaves zero room for execution error or a slowdown in hyperscaler AI infrastructure spending."
Broadcom (AVGO) is effectively the 'picks and shovels' play of the AI gold rush, but investors are paying a steep price for that stability. An 81x trailing P/E suggests the market has already priced in perfection, assuming that the 140% growth in custom ASICs is a permanent baseline rather than a cyclical surge. While the 77% gross margins are elite, they are highly sensitive to the concentration risk of its top-tier hyperscaler clients like Google and Meta. If these tech giants pivot toward internalizing more silicon design or if capex budgets face a 'cooling-off' period, AVGO’s premium valuation will face a brutal re-rating. It is a quality compounder, but currently trading at a 'growth-at-any-price' multiple.
The bull case ignores that AVGO is essentially an infrastructure utility; if AI demand persists, their custom silicon and networking moats are so wide that they can sustain these margins regardless of broader sector volatility.
"Broadcom’s premium valuation is not adequately priced for a potentially cyclical AI capex recovery, given ambitious revenue targets and competition that could erode margins."
Broadcom benefits from AI tailwinds in both custom chips and data-center networking, but the bull case hinges on an extraordinary growth trajectory and a generous multiple. The article highlights a $100B AI-chip revenue target by 2027, implying a multi-year CAGR that may be unrealistically steep given intensifying competition (Nvidia, AMD, in-house accelerators) and potential customer risk concentration (Google, Alphabet). Even with 77% gross margins and a dividend, the stock trades near MarketCap-uber-valuation levels (P/E around 81) and could face margin pressure if AI spend moderates or supply/demand cycles normalize. The absence of margin resilience under a cyclical capex backdrop is glossed over.
The bulls could be right if AI spend remains robust and Broadcom secures durable, high-margin clients; a long-term Google/TPU collaboration could sustain demand and pricing power. However, that hinges on an uninterrupted mega-cycle that may not materialize.
"Networking mix shift may compound rather than diversify client concentration risk."
Claude correctly notes the unquantified concentration but misses how Broadcom's networking push to 40% of AI sales could actually amplify single-client exposure if those same hyperscalers slow Ethernet adoption. The 60% Q1 networking growth already embeds that bet; any delay in the mix shift would hit both revenue and the 77% margin assumption simultaneously, a second-order linkage the valuation at 81x does not appear to discount.
"Concentration risk isn't just revenue volatility—it's margin compression when customers have leverage across multiple product lines."
Grok's networking-concentration linkage is sharp, but it inverts the actual risk. If hyperscalers slow *overall* capex, both custom chips and networking decline together—that's correlated downside, not amplified. The real vulnerability: Broadcom's 77% margins assume pricing power *within* each segment. If Google negotiates harder on networking while shifting volume to in-house TPUs, margins compress on both fronts simultaneously. The 81x multiple doesn't price in a margin squeeze, only revenue miss.
"Broadcom's valuation is artificially propped up by non-GAAP software accounting that masks the underlying hardware cyclicality and margin risks."
Claude, you hit the margin compression risk perfectly, but we are missing the 'VMware' elephant in the room. Broadcom’s valuation isn't just an AI play; it’s an aggressive software-margin expansion story post-acquisition. The 81x P/E is inflated by integration costs and non-GAAP accounting adjustments that mask core hardware cyclicality. If networking growth decelerates, the market will stop valuing AVGO as a high-growth AI darling and revert to pricing it as a mature, low-growth semiconductor utility.
"The macro AI capex cycle risk is the real pressure on margins and valuation, not VMware integration."
Gemini, the VMware angle may be material but it’s not the core risk. The bigger, underappreciated flaw is the macro capex cycle for AI: if hyperscalers pause or slow, Broadcom’s dual engines - custom chips and networking - die together, and margins will compress as pricing power wanes. The 81x trailing P/E already encodes near-perfection; any pullback in AI spend or a broader semiconductor downturn could trigger a sharp multiple re-rating regardless of VMware synergies.
Panelists agree that Broadcom's strong AI chip and networking growth, high gross margins, and Google TPU deal provide near-term momentum, but the stock's high valuation (81x trailing P/E) may not be sustainable given potential risks such as customer concentration, margin compression, and a slowdown in hyperscaler capex.
None explicitly stated.
Margin compression due to customer negotiations and a slowdown in hyperscaler capex, as well as customer concentration risk.