Broadcom stock is heading into one of its most important earnings reports of 2026 as investors look for proof that explosive demand for custom AI accelerators and networking chips can justify AVGO’s roughly $1.8 trillion valuation. Broadcom reports fiscal third-quarter results after the market closes on Wednesday, September 2, with Wall Street focused on approximately $29.4 billion of company revenue, $16 billion of AI semiconductor sales and whether management can reinforce its ambitious path toward more than $100 billion in annual AI revenue.
The stakes are unusually high because Broadcom is no longer being valued as a conventional diversified semiconductor company. Investors increasingly see it as one of the most important alternatives and complements to Nvidia inside the AI infrastructure boom, thanks to custom accelerators built for hyperscale customers and networking products required to connect enormous AI clusters. That opportunity has transformed the company’s growth profile, but it has also raised expectations to a level where merely beating quarterly estimates may not be enough to send Broadcom stock higher.
Stock Needs More Than Another Earnings Beat
Broadcom closed September 1 at $369.68, down 0.18% for the session. Shares were still up roughly 6.8% for 2026 and about 24% over the previous year, but they had fallen more than 23% over three months and sat well below their 52-week high of $495. Broadcom’s market capitalization stood near $1.76 trillion, leaving virtually no room for investors to treat the upcoming report as routine.
That recent weakness is important because it tells investors something about the market’s expectations. Broadcom has delivered extraordinary growth, yet AVGO has dramatically underperformed the broader semiconductor index during 2026. MarketWatch reported that Broadcom was up only about 6.7% year to date compared with a roughly 60% gain in the PHLX Semiconductor Index, as investors questioned competition in custom AI chips and sought more clarity on management’s longer-term AI targets.
In other words, strong numbers are already assumed. The question is whether CEO Hock Tan can deliver another increase in the market’s expectations.
Related articles
AI Revenue Is the Number Wall Street Cannot Ignore
Broadcom’s AI semiconductor business has expanded at a staggering pace. During the fiscal second quarter, AI semiconductor revenue reached $10.8 billion, up 143% from a year earlier and ahead of management’s forecast. Broadcom said the growth was driven by accelerating demand for both custom AI accelerators and AI networking products.
For the third quarter, management forecast AI semiconductor revenue of approximately $16 billion, representing year-over-year growth of more than 200%. That means AI revenue alone could account for more than half of Broadcom’s projected total quarterly sales of around $29.4 billion. The shift illustrates how rapidly artificial intelligence has changed the company’s revenue mix.
Only six months earlier, Broadcom had reported $8.4 billion of fiscal first-quarter AI revenue, up 106% year over year. Moving from $8.4 billion in Q1 to a projected $16 billion in Q3 would represent an extraordinary acceleration within just two quarters.
That growth is precisely why investors have assigned Broadcom such a premium valuation. It is also why anything short of another aggressive outlook could disappoint the market.
The $100 Billion AI Target Is Becoming the Real Catalyst
Broadcom’s quarterly numbers may dominate the headlines, but longer-term commentary could matter more for AVGO stock.
After its previous results, Hock Tan reiterated expectations for Broadcom to generate more than $100 billion of AI semiconductor revenue in fiscal 2027. Reuters reported that management also expects to support more than 10 gigawatts of AI computing deployments by that point.
Wall Street is now looking for more evidence behind that extraordinary number. Investors want to know how many hyperscale customers are contributing, how quickly new custom accelerators are ramping and whether Broadcom can maintain its dominant position as large cloud companies increasingly design their own AI silicon.
MarketWatch reported that JPMorgan analyst Harlan Sur believes the earnings call could provide greater clarity on the durability of Broadcom’s AI growth and its path toward the $100 billion-plus target. The concern is not that AI spending is disappearing; rather, it is whether competitors can capture a larger share of the custom-chip opportunity.
That distinction is crucial. Broadcom does not need Nvidia to lose for AVGO to win. It needs hyperscalers to continue building custom AI infrastructure at enormous scale.
Nvidia Just Raised the Bar for Broadcom Earnings
Broadcom’s report comes immediately after Nvidia delivered another powerful signal that AI infrastructure spending remains intact. Nvidia recently forecast roughly 70% revenue growth for its next fiscal year, significantly above prior Wall Street expectations, while data-center sales continued to surge.
The results triggered a broad semiconductor rally and helped reassure investors that hyperscalers are not pulling back from AI spending despite concerns about the enormous capital commitments required. Broadcom was among the companies benefiting from the renewed optimism.
But Nvidia’s performance also creates a tougher comparison.
Investors have now seen Nvidia deliver enormous growth and a powerful longer-term outlook. Broadcom must convince Wall Street that its custom-chip and networking businesses are participating in the same AI spending cycle without suffering a meaningful slowdown.
A merely acceptable report could therefore look disappointing against the backdrop of Nvidia’s latest forecast.
Custom AI Chips Could Be Broadcom’s Biggest Opportunity
Broadcom occupies a different position in AI infrastructure than Nvidia. Nvidia primarily sells general-purpose GPUs and complete AI computing platforms, while Broadcom works with some of the largest cloud providers to design highly customized accelerators optimized for particular workloads.
These application-specific integrated circuits, or ASICs, can be attractive to hyperscalers because they may offer better cost efficiency and power consumption for workloads running at enormous scale. Alphabet has historically been one of Broadcom’s most important custom-silicon relationships, and the broader trend toward internally designed chips could expand the addressable market considerably.
Yet competition is rising.
Marvell has become more aggressive in custom silicon and recently strengthened its position through a major Google-related agreement. Reuters reported that Marvell expects custom-chip revenue to more than double next year, although meaningful contribution from one major new agreement may not arrive until later years.
Investors are also monitoring reports that large hyperscalers may diversify their design partners. MarketWatch noted concerns that Google could use additional semiconductor partners alongside Broadcom. JPMorgan nevertheless remains confident in Broadcom’s relationship with Alphabet, citing the depth and duration of their partnership.
For Broadcom shareholders, customer diversification matters almost as much as overall AI growth.
Networking Is Becoming Just as Important
Broadcom’s AI story is not limited to custom compute chips.
As AI clusters grow from thousands to potentially hundreds of thousands or even millions of accelerators, moving data efficiently between chips becomes an increasingly serious technical challenge. Broadcom sells networking silicon and connectivity products that help solve that problem, giving it exposure to AI infrastructure even when the accelerator itself comes from another supplier.
That could become a critical advantage as AI systems scale.
Industry analysts increasingly argue that networking and connectivity could become the next major bottleneck in data centers. MarketWatch reported this week that Deutsche Bank sees networking as a major next phase of AI infrastructure spending because massive clusters require increasingly sophisticated connectivity.
Broadcom therefore has two ways to benefit from the same AI capital-expenditure cycle: custom compute and the networking infrastructure connecting that compute.
That combination is one reason Wall Street has become comfortable placing AVGO in the same conversation as Nvidia when discussing AI beneficiaries.
Margins Could Decide How Investors React
Revenue growth will attract most of the attention, but profitability could determine whether Broadcom stock rallies after the report.
Broadcom guided for fiscal third-quarter non-GAAP operating income equal to approximately 67% of revenue and adjusted EBITDA of roughly 68%. In Q2, adjusted EBITDA reached $15.24 billion, or approximately 69% of revenue, while free cash flow totaled $10.26 billion, representing 46% of revenue.
Those are extraordinary profitability levels for a company growing this rapidly.
Broadcom’s SEC filing showed a GAAP gross margin of 69% in the fiscal second quarter, up from 68% a year earlier. The company attributed much of the improvement to strong demand for AI-related semiconductor products.
The danger is that AI hardware supply constraints and rising component costs could eventually pressure those margins. Nvidia has already warned that memory shortages and higher costs are weighing on profitability, with its margin outlook moving lower as next-generation products ramp.
If Broadcom can maintain margins near current levels while AI revenue grows more than 200%, the earnings leverage could remain enormous.
VMware Remains Quiet Profit Engine
The AI semiconductor story receives most of the attention, but Broadcom’s infrastructure-software segment remains essential to its financial profile.
The company’s SEC filing showed that infrastructure-software revenue increased year over year primarily because of strong demand for VMware Cloud Foundation. Broadcom has spent several years restructuring VMware’s product portfolio, pricing and customer relationships after completing its acquisition.
That recurring software revenue gives Broadcom something many semiconductor companies lack: a large, relatively predictable source of cash flow outside the volatile chip cycle.
For investors, VMware matters because software margins help support Broadcom’s enormous consolidated profitability while AI semiconductors deliver the headline growth. Any evidence of slowing VMware demand or customer pushback would therefore deserve attention, even if semiconductor revenue beats expectations.
Is Broadcom Stock a Buy Before Earnings?
The bullish case remains powerful. Broadcom is producing triple-digit AI semiconductor growth, generating more than $10 billion of quarterly free cash flow and participating in both custom accelerators and AI networking. Management continues to target more than $100 billion of AI revenue in fiscal 2027, while Nvidia’s latest outlook suggests the underlying AI infrastructure cycle remains exceptionally strong.
The risk is valuation and expectations.
At $369.68, Broadcom trades at roughly 61 times trailing earnings, according to MarketWatch. A company worth approximately $1.76 trillion must deliver enormous future cash flows to justify that valuation, and the recent 23% three-month decline shows investors are willing to punish even small cracks in the AI narrative.
For investors buying Broadcom before earnings, the trade is therefore not simply a bet that the company will beat estimates. It is a bet that management will raise or reinforce expectations enough to satisfy one of the most demanding investor bases in technology.
Outlook: AI Forecast Could Move the Entire Chip Sector
Broadcom reports after Wednesday’s closing bell, with its earnings call beginning at 5 p.m. Eastern Time. The critical numbers will be Q3 revenue, AI semiconductor sales, margins and fourth-quarter guidance, but commentary on fiscal 2027 could ultimately produce the biggest stock move.
Investors should listen closely for updates on the $100 billion-plus AI target, additional hyperscale customers, custom accelerator ramps, networking demand and VMware profitability.
The macro backdrop adds another layer of risk. U.S. stocks began September under pressure as Treasury yields climbed and investors increased expectations for another Federal Reserve rate hike, creating a more difficult environment for richly valued technology stocks.
Broadcom has already proved that AI demand is exploding.
Now the company must prove that growth can remain extraordinary enough to justify a nearly $1.8 trillion valuation—and Wednesday night could reveal whether Wall Street thinks the AI boom still has much further to run.
