By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panel is largely bearish on semiconductor stocks, warning that current valuations (AMD at 69x, Broadcom at 40x) leave little room for error and are pricing in sustained, flawless execution for years. They caution that a modest growth deceleration or disappointing results could trigger multiple compression.
Risk: Disappointing results or a modest growth deceleration could trigger multiple compression, leading to significant valuation drops.
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 →
AMD's data center revenue grew 57% in the first quarter of 2026.
Qualcomm jumped after reportedly landing ByteDance as an AI data-center customer.
Broadcom reports its next quarterly results on June 3.
Advanced Micro Devices just had one of its best months in two decades. The stock soared more than 40% in May to a record high, as investors poured back into chipmakers on relentless demand for AI hardware. And it wasn't alone. Smartphone-chip designer Qualcomm climbed about 40% to a record of its own, while the broader Philadelphia Semiconductor Index gained more than 60% for the year, far outpacing the S&P 500.
After a stretch like that, investors heading into summer face a fair question: can the run keep going, or have the stocks finally outpaced the businesses underneath them?
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 »
It turns out that, for many of the chip companies behind these stocks, the underlying businesses do seem to be living up to their stock gains.
AMD, for instance, gave investors plenty to like when it reported first-quarter results in early May. Revenue rose 38% year over year to $10.3 billion, led by its data center segment, which grew 57% to a record $5.8 billion on demand for the company's EPYC server processors and Instinct GPUs. Its non-GAAP (adjusted) earnings per share grew 43%, and management guided for about $11.2 billion in second-quarter revenue -- a figure that implies about 46% year-over-year growth.
"These results mark a clear inflection in our growth trajectory and a structural shift in our business," said AMD Chair and CEO Lisa Su in the company's first-quarter earnings call.
Custom-chip designer Broadcom (NASDAQ: AVGO) is catching the same wave, but from a different seat. In its fiscal first quarter of 2026 (the period ended Feb. 1, 2026), revenue rose 29% to a record $19.3 billion, and its AI revenue more than doubled, up 106% to $8.4 billion, on demand for the custom accelerators and networking chips it builds for the largest cloud companies. It now counts six major custom-AI-chip customers, guided for about $22 billion in fiscal second-quarter revenue, up 47%, and reports those results on June 3.
Qualcomm's May surge, however, was a different story. Its stock's gain was based on more speculative news. In late May, Bloomberg reported that the company had reached a deal to supply TikTok owner ByteDance with millions of custom chips for AI data centers -- one of its first big customers for a product line meant to carry it beyond smartphones. Neither company confirmed the agreement, but the stock jumped on the news. In addition, investors seem to be realizing that Qualcomm could be a beneficiary of AI as it makes its way into the physical world.
But have investors who haven't bought into these growth stories already now missed out?
As of this writing, AMD trades at a forward price-to-earnings ratio of about 69 and Broadcom trades at more than 40 times forward earnings -- multiples that assume the AI build-out keeps compounding for years without a stumble.
Qualcomm's stock is much cheaper, at around 21 times forward earnings. But there's a reason for its comparative discount. Its core business is actually shrinking. Fiscal second-quarter revenue (the period ended March 29) fell 3% year over year as handset chip sales dropped 13%, and adjusted earnings per share fell about 7%. Additionally, management then guided for a sequential step-down in the current quarter. So what's helping Qualcomm stock? Investors seem to be betting that the reported ByteDance deal pays off and that its chip sales will surge as AI moves into devices.
Overall, however, there are good reasons for chip stocks' big gains in May, from Broadcom's success with custom chips for major tech companies to Intel's turnaround progress; the catalysts are mostly undeniable.
But Broadcom's June 3 report is arguably the next real test. A strong quarter and confident outlook could carry the group's momentum into summer; any hint of cooling demand, however, could do the opposite. Either way, the businesses themselves look healthy enough to keep growing even if the stocks take a breather.
For now, I don't think the biggest risk for investors is that AI demand suddenly dries up. It's that, at these valuations, merely good results may no longer be good enough. After a May like this one, a little patience may serve investors better than chasing the rally higher.
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 $462,983! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,375,447!
Now, it’s worth noting Stock Advisor’s total average return is 995% — a market-crushing outperformance compared to 212% 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 June 2, 2026. *
Daniel Sparks has and his clients have no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Advanced Micro Devices, Broadcom, Intel, and Qualcomm. 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
"AMD and Broadcom have earned their gains operationally, but Qualcomm's 40% rally on an unconfirmed rumor while its core business shrinks is the canary—these valuations leave zero margin for error."
The article conflates two very different stories. AMD and Broadcom show real operational leverage—57% and 106% segment growth respectively, with guidance implying sustained momentum. But AMD at 69x forward P/E and Broadcom at 40x are pricing in flawless execution for years. Qualcomm is the real tell: its core smartphone business is *shrinking* (down 3% YoY, guidance for sequential decline), and the entire 40% May rally rests on an unconfirmed ByteDance rumor. The article acknowledges this but buries it. If Broadcom's June 3 results disappoint even modestly—say, 40% guidance instead of 47%—these valuations compress hard. The article's closing line is doing heavy lifting: 'merely good results may no longer be good enough.' That's not reassuring; it's a warning dressed as wisdom.
AI capex is genuinely structural and may justify elevated multiples if TAM expands faster than consensus assumes; Broadcom's custom-chip moat with six hyperscaler customers is defensible and could sustain 40x+ multiples if execution remains flawless through 2027.
"Valuations already embed multi-year perfection, so merely healthy results on June 3 may trigger de-rating rather than continuation."
Semiconductor stocks rode genuine Q1 beats—AMD data-center revenue +57% to $5.8B, Broadcom AI sales +106%—yet now price in flawless execution for years. AMD’s 69x forward P/E and Broadcom’s 40x+ leave scant margin for the June 3 print or any sequential slowdown in hyperscaler capex. Qualcomm’s 40% May spike rests on an unconfirmed ByteDance rumor rather than core handset trends that are still contracting. The overlooked risk is not demand evaporation but a modest growth deceleration that still triggers multiple compression after such a parabolic run.
Even in-line Broadcom results could trigger another leg higher if management signals further AI customer wins or raises full-year guidance, as the market has repeatedly rewarded any incremental visibility into the custom-ASIC pipeline.
"At current forward valuations, semiconductor stocks have moved from growth plays to high-risk bets that require flawless execution, making them vulnerable to even minor earnings misses."
The semiconductor rally is shifting from broad enthusiasm to a 'show me' phase. While AMD and Broadcom are delivering legitimate growth, the 69x forward P/E for AMD is pricing in perfection, leaving zero margin for execution errors. The real story is the divergence: Broadcom is a play on infrastructure 'picks and shovels,' whereas the Qualcomm/ByteDance narrative is speculative, relying on unconfirmed deals to offset a shrinking core handset business. Investors are ignoring the cyclicality of the semiconductor industry; when hyperscalers eventually optimize their AI spend, the current double-digit revenue growth rates will face a harsh reality check. We are at peak valuation, not peak earnings.
If AI infrastructure becomes a permanent, utility-like capital expenditure for the global economy, these 'expensive' multiples are actually justified by a structural shift in long-term margin profiles.
"The market's strength hinges on durable AI-driven demand lasting into the next several years, but near-term missteps or a moderation in cloud capex could trigger rapid multiple compression."
May's surge in AMD (+~40%), Broadcom (+~40% to record highs), and Qualcomm (+~40%) reflects AI datacenter demand and cloud capex, with Broadcom guiding to about $22B in Q2 revenue. The narrative lays the groundwork for multi-year AI infrastructure growth. Yet valuations are extreme: AMD ~69x forward earnings, Broadcom ~40x, Qualcomm ~21x, implying the market is pricing sustained AI expansion well beyond profitability visibility. ByteDance's rumored deal for QCOM remains unconfirmed, and Qualcomm's core handset business has cooled. If AI demand cools, or Broadcom's June print disappoints, multiple compression could hit fast. The real test is durability of AI-driven pricing power, not a single quarter.
Bearish counterpoint: The rally is already pricing in years of uninterrupted AI capex; a data-point miss or a macro slowdown could crush valuations quickly, especially for AMD and Broadcom given their lofty multiples.
"The panel is debating whether valuations are stretched, but nobody has modeled the specific growth-rate threshold where multiple compression begins—that's the real vulnerability."
Claude and Gemini both flag the 69x AMD multiple as pricing perfection, but neither quantifies what 'modest' deceleration means. If AMD's data-center growth drops from 57% to 35% YoY—still exceptional—does the multiple compress to 45x or 30x? That gap determines whether June 3 is a speed bump or a cliff. Broadcom's custom-chip moat is real, but Grok's point about in-line results triggering further upside assumes management can *always* raise guidance. That's not cyclical; that's a perpetual motion machine.
"June 3 hinges more on forward commentary than the headline growth rate."
Claude flags the risk of Broadcom failing to perpetually raise guidance, yet misses how lengthening AI training cycles could let even a 35% AMD data-center print avoid multiple compression if it includes 2026 visibility. That distinction matters because the market has rewarded incremental pipeline color more than raw growth rates in prior quarters. June 3 may hinge on forward commentary rather than the headline number itself.
"The market is ignoring the impact of a higher-for-longer discount rate on lofty semiconductor valuation multiples."
Grok, your focus on 2026 visibility ignores the 'law of large numbers' hitting Broadcom. Even with custom-ASIC wins, Broadcom’s revenue base is massive; maintaining 40% growth requires exponential, not linear, hyperscaler spend. If Broadcom prints 'in-line' but provides conservative 2025 commentary, the market will punish the 40x multiple instantly. We are ignoring the macro headwind: if the Fed holds rates higher for longer, the discount rate on these long-duration growth assets will force a violent multiple contraction.
"Inline 2026 visibility isn’t a free pass; even 35% AMD data-center growth could trigger multiple compression if AI capex cools or rates stay higher."
Grok's line that inline results could still lift shares thanks to 2026 visibility ignores a harsher math: even 35% data-center growth in AMD may not sustain a 40x+ multiple if AI capex decelerates or rates stay high. The risk is a two-step re-rating: first lower growth surprises, then multiple compression. June 3 guidance matters, but duration risk and cyclicality unchanged. The 'visibility' thesis needs a clear path to sustainable margin expansion.
The panel is largely bearish on semiconductor stocks, warning that current valuations (AMD at 69x, Broadcom at 40x) leave little room for error and are pricing in sustained, flawless execution for years. They caution that a modest growth deceleration or disappointing results could trigger multiple compression.
None explicitly stated.
Disappointing results or a modest growth deceleration could trigger multiple compression, leading to significant valuation drops.