
• ASML's CEO confirmed that the first chips produced on High-NA machines are expected within months, covering both memory and logic. This dismantles the bear case that High-NA was still years from maturity.
• Five of the largest US hyperscalers — Amazon, Microsoft, Alphabet, Meta, and Oracle — are now collectively guiding for approximately USD 750 billion in capital expenditure in 2026, up from roughly USD 680 billion previously.
• ASML remains the only company in the world capable of producing EUV lithography machines commercially. No credible competitor is close, and with High-NA now entering volume production the technology gap is likely to become even wider than before.
• We continue to hold a positive view on ASML with a target price of USD 1,750, representing an upside potential of 12.8% (as of 21 May 2026).
ASML confirmed that the first chips made with High-NA EUV machines will be delivered within months.
On 19 May 2026, ASML (NASDAQ:ASML) CEO Christophe Fouquet told attendees at the IMEC conference in Antwerp: “In the next few months, we will be looking at the first few products wherever, in memory, in logic, being exposed on the High-NA system.” The significance of this is straightforward: High-NA is no longer a technology being evaluated in customer R&D labs. It is starting to produce real chips.
So far, the strongest pushback on High-NA EUV has been on price. TSMC (ASML’s largest customer) said publicly in recent weeks that these machines, priced at up to USD 400 million each were too expensive.
But cost does not tell the whole story.
Fouquet addressed this directly at the same conference. High-NA EUV is engineered to reduce the number of patterning passes needed at advanced nodes, which lowers cost per chip over time. The economics at scale look very different from those of early-stage R&D deployment.
More tellingly, Intel, TSMC, Samsung, and SK Hynix have all received and installed High-NA units, with several placing additional orders. If the cost hurdle were truly prohibitive, those commitments would not exist. We believe that TSMC's conservatism on High-NA EUV is a question of timing, not direction. As chip designs become more complex at each successive node, the cost economics of High-NA improve and the case for adoption becomes harder to resist.
The AI spending cycle just got bigger again
Five of the largest US hyperscalers (Amazon, Microsoft, Alphabet, Meta, and Oracle) are now collectively expected to spend USD 750 billion on capital expenditure in 2026, up materially from prior projections of around USD 680 billion, and more than 73% above 2025 levels (Figure 1). That is more than USD 300 billion in additional spending in a single year, a clear signal that the AI megatrend is not just alive, but intensifying.
Figure 1: US hyperscalers have ramped up their capital spending plans once more

This capital does not remain at the hyperscaler level. It flows downstream into leading-edge chips, new fabs, and the lithography equipment required to manufacture those chips. Last month, TSMC raised its capex outlook to the upper end of its USD 52–56 billion 2026 guidance range, roughly 32% above 2025 levels.
TSMC is currently the one notable holdout on High-NA EUV adoption, preferring to extend the life of existing low-NA EUV machines through innovative chip design rather than commit to machines priced at up to USD 400 million each. That conservative stance is reflected in its current capex guidance. But as we noted earlier, the question is one of timing, not direction. As nodes advance and the limits of low-NA EUV are reached, High-NA adoption becomes increasingly difficult to defer.
When TSMC does move, the capex implications are significant. High-NA machines carry roughly double the price tag of existing EUV systems. Therefore, a meaningful ramp from the world's largest foundry would push both TSMC's own capex and ASML's order book materially higher than what current guidance already implies.
Every dollar of hyperscaler capex that flows into leading-edge chips eventually lands on ASML's order book because there is nowhere else for it to go.
ASML's monopoly safeguards long-term earnings resilience
The demand case is compelling on its own. But what makes ASML truly difficult to replicate as an investment is its position on the supply side. No other company can do what it does.
Till this day, ASML remains the only company capable of producing EUV lithography machines commercially, while also commanding approximately 90% of the high-end DUV market. No credible competitor is close. China’s domestic lithography champion is still working to reliably produce 28nm DUV tools, a technology ASML commercialised in the late 2000s. Japan’s Nikon and Canon abandoned EUV development over a decade ago.
With High-NA now entering production, the technology gap is set to widen even further. Every machine that ASML ships adds to the installed base, and each machine can generate service income for several decades. As the High-NA fleet grows, so does this compounding stream of recurring, high-margin income. Think of it like a subscription that lasts several decades per unit and gets larger with every new machine shipped.
Related Article: ASML powers ahead as customer upgrades reinforce 20% upside potential
While the upside has narrowed following the recent rally, our long-term conviction in ASML is unchanged. If anything, TSMC's eventual adoption of High-NA (which is not yet reflected in current estimates) represents a meaningful catalyst that could push both earnings and the share price materially higher from here.
ASML is currently trading at approximately 28.3X 2028 estimated earnings, below our assigned fair multiple of 32X. This translates to an upside potential of 12.8% (as of 21 May 2026) and a target price of USD 1,750.
To reiterate, ASML (NASDAQ:ASML) remains one of the most compelling long-term holdings in the semiconductor space. High-NA chips are arriving, the AI capex cycle keeps expanding, and the monopoly is deepening. For existing holders, we recommend staying invested. For investors without exposure, any near-term pullback, whether driven by macro noise or the ongoing cost debate around High-NA should be viewed as an opportunity to build positions in this stock.
Table 1: ASML is expected to see robust earnings growth in the coming years
|
|
2025 |
2026E |
2027E |
2028E |
|
EPS (EUR) |
26.29 |
31.20 |
40.90 |
47.00 |
|
EPS growth |
23.83% |
18.68% |
31.09% |
14.91% |
|
PE Ratio |
35.05 |
42.72 |
32.59 |
28.36 |
|
Upside Potential (based on 32X multiple) |
- |
- |
- |
12.83% |
|
Source: Bloomberg Finance L.P., iFAST Estimates |
||||
|
Data as of 21 May 2026 |
||||

Declaration:
For specific disclosure, at the time of publication of this report, IFPL (via its connected and associated entities) holds a NIL position in the abovementioned securities. The analyst who produced this report holds a position in ASML.
This research report was prepared with the assistance of artificial intelligence (AI) tools. iFAST Financial Pte Ltd does not rely exclusively on AI for content generation; the content of this report – including all investment theses, ratings, price targets and conclusions – has been independently reviewed and verified by the research analyst(s) to ensure accuracy and professional integrity.
