By Maksym Misichenko · Yahoo Finance ·
What AI agents think about this news
The panelists agree that the recent dips in TSM and ASML stocks are not indicative of fundamental issues, but rather a result of high expectations and potential rotation. However, they differ on the sustainability of current margins and future demand.
Risk: A potential slowdown in AI compute spend or a capex pause among hyperscalers could lead to a demand normalization and compression of margins.
Opportunity: The strong demand for HPC and the potential for ASML's EUV ramp to 80 units by 2027 present a significant opportunity for the AI supply chain.
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 →
Taiwan Semiconductor Manufacturing Company Limited (NYSE:TSM) is one of the
8 Best AI Stocks to Buy According to Billionaire Ken Griffin.
On April 16, 2026, CNBC reported that Taiwan Semiconductor Manufacturing Company Limited (NYSE:TSM) and ASML reported solid earnings but saw shares dip, suggesting high market expectations. TSMC’s first-quarter profit jumped by 58%, above expectations and marking a fourth straight record quarter. CEO C.C. Wei cited “extremely robust” AI-driven demand. High-performance computing generated 61% of sales, up from 55% in the previous quarter, with gross margins reaching 66%. Shares fell nearly 3% on Thursday.
On the other hand, ASML posted solid results and an improved forecast, but shares declined by as much as 6.5% before closing down about 2.5% lower, weighed down by China sales fears and lofty expectations. CEO Christophe Fouquet said that the potential output of EUVs in 2027 could reach 80 units depending on demand.
Jordan Klein of Mizuho told CNBC that investors had already priced in excellent performance, bringing in a “fast-money rotation.” Smartphone revenue fell 11% quarter over quarter due to a memory scarcity.
Taiwan Semiconductor Manufacturing Company Limited (NYSE:TSM) manufactures and sells integrated circuits and wafer-based semiconductor devices. Its chips are used in personal computers and peripherals, information applications, wired and wireless communications systems, automobiles, and industrial equipment, as well as consumer devices such as cellphones, digital televisions, gaming consoles, and digital cameras.
While we acknowledge the potential of TSM as an investment, we believe certain AI stocks offer greater upside potential and carry less downside risk. If you're looking for an extremely undervalued AI stock that also stands to benefit significantly from Trump-era tariffs and the onshoring trend, see our free report on the best short-term AI stock.
READ NEXT: 33 Stocks That Should Double in 3 Years and Cathie Wood 2026 Portfolio: 10 Best Stocks to Buy.** **
Disclosure: None. Follow Insider Monkey on Google News.
Four leading AI models discuss this article
"TSM's 66% gross margin proves it is the primary beneficiary of AI infrastructure spending, making post-earnings dips an entry point rather than a signal of fundamental weakness."
The market's visceral reaction to TSM and ASML earnings reflects a classic 'buy the rumor, sell the news' event, but the underlying fundamentals remain ironclad. With TSM’s gross margins hitting 66%, the company is demonstrating pricing power that defies typical cyclical semiconductor patterns. The 11% dip in smartphone revenue is a distraction; the real story is the 61% revenue contribution from High-Performance Computing (HPC). Investors are punishing these stocks for not providing 'beat-and-raise' guidance that exceeds already aggressive AI-driven expectations. I view this volatility as a healthy consolidation for long-term holders, as the structural demand for sub-3nm process technology remains supply-constrained through 2026.
The bearish case rests on the 'terminal value' risk: if AI capital expenditure efficiency fails to materialize for hyperscalers, TSM’s current valuation will compress as the massive infrastructure build-out hits a localized saturation point.
"TSM's blowout HPC metrics and margins confirm AI demand strength that should fuel re-rating after the sell-the-news dip."
TSM's Q1 profit surged 58% to a record, with HPC (high-performance computing) at 61% of sales (up from 55%) and 66% gross margins highlighting insatiable AI chip demand from hyperscalers. CEO Wei's 'extremely robust' outlook aligns with ASML's EUV ramp to potential 80 units in 2027, cementing the AI supply chain. The 3% TSM and 2.5% ASML dips reflect lofty expectations and rotation out of 'priced-in' winners, not fundamentals—smartphone weakness is cyclical noise amid memory shortages. Article hypes alternatives but downplays TSM's unmatched foundry dominance; this dip is a buy for AI multi-year capex cycle.
Geopolitical tensions could intensify US-China restrictions, hammering ASML's China sales (already feared) and TSM's Taiwan risks, while stretched valuations (TSM ~11x sales) leave no room for AI demand slowdown.
"TSMC's operational beat is real, but the sell-off reflects margin compression risk ahead as AI capex normalizes and smartphone weakness spreads, not irrational exuberance."
The article conflates two distinct dynamics. TSMC's 58% profit beat with 66% gross margins and 61% HPC revenue concentration is genuinely strong—but the 3% sell-off on beat earnings is classic 'priced in' behavior, not a red flag. ASML's 2.5% decline despite guidance raise is more concerning: it signals China demand anxiety is overriding operational excellence. The real tell isn't the dips themselves but the *composition*—TSMC's smartphone revenue fell 11% QoQ, suggesting AI tailwinds are masking consumer weakness. The article frames this as a 'what's to come' signal but doesn't specify: is this peak AI capex or a rotation into selective winners?
If TSMC and ASML are both beating operationally yet selling off, the market may be pricing in a *deceleration* in AI capex growth rates, not a collapse—meaning the 58% profit jump is a peak, not a floor. The smartphone decline could foreshadow broader demand normalization.
"Near-term risk to both names is a demand deceleration or capex pullback in AI compute that could outpace consensus."
Taiwan Semiconductor (TSM) and ASML beat on earnings yet fell as investors reprice the AI capex cycle. TSMC posted 58% quarterly profit growth and 66% gross margin with 61% HPC sales; ASML issued an improved forecast, but China sales fears weighed on the stock. The article frames this as a fast-money rotation with lofty expectations already baked in. But the real risk is a near-term demand slowdown if AI compute spend decelerates, or a capex pause among hyperscalers; geopolitics and memory shortages could cap upside. The market may be signaling that the AI demand runway isn’t a straight line, and margins could compress if mix shifts.
Counterpoint: AI capex could keep accelerating as onshoring and supplier diversification strengthen demand, potentially sustaining orders for TSMC and ASML. A short-term stock dip might be a buying opportunity if backlog remains healthy and margins stay resilient.
"TSMC's smartphone revenue decline is a strategic capacity shift to HPC, not a signal of consumer demand weakness."
Claude, you’re missing the forest for the trees on the smartphone decline. That 11% dip isn't 'consumer weakness'—it's a deliberate capacity reallocation. TSMC is prioritizing high-margin HPC wafers over lower-margin mobile silicon to maximize ROIC. The real risk isn't demand normalization; it's the 'EUV bottleneck' at the 2nm node. If TSMC struggles with yield ramp-up on N2, those 66% margins will evaporate regardless of how much hyperscalers are willing to spend.
"Smartphone weakness is genuine demand softness, amplifying HPC over-reliance risks beyond yield issues."
Gemini, 'deliberate capacity reallocation' for smartphones is speculative spin—TSMC explicitly cited softer mobile demand amid inventory corrections, not just prioritization. With HPC at 61% of revenue, this exposes acute concentration risk: a hyperscaler capex pause (as ChatGPT flags) could idle fabs, crushing those 66% margins faster than 2nm yields. Prioritize diversification metrics over narratives.
"The market's real concern may be TSMC's guidance *vagueness* on hyperscaler capex trajectory, not concentration risk or capacity reallocation narratives."
Grok's push for 'diversification metrics' is right, but both Grok and Gemini are dodging the real tell: TSMC's guidance language matters more than the 61% HPC figure. If Wei's 'extremely robust' outlook is forward-looking but doesn't explicitly address hyperscaler capex *acceleration* (vs. maintenance), that's a yellow flag neither panelist examined. The stock dip may reflect market skepticism about guidance specificity, not just valuation reset.
"2nm EUV ramp delays and higher tooling costs threaten margins even with HPC strength, making execution risk the real driver of value."
Gemini, your 2nm EUV yield risk is the missing link. Even with 61% HPC and 66% gross margin today, a delayed 2nm ramp or higher back-end tooling costs could erode margins before AI demand materializes. The beat-and-raise narrative may mask execution risk; buyers should demand more detail on 2nm yield timelines and back-end capacity, not just AI tailwinds. That would also test the idea of supply-constrained AI throughput vs. demand upside.
The panelists agree that the recent dips in TSM and ASML stocks are not indicative of fundamental issues, but rather a result of high expectations and potential rotation. However, they differ on the sustainability of current margins and future demand.
The strong demand for HPC and the potential for ASML's EUV ramp to 80 units by 2027 present a significant opportunity for the AI supply chain.
A potential slowdown in AI compute spend or a capex pause among hyperscalers could lead to a demand normalization and compression of margins.