They are replacing 30% of their revenue with a 'strategic' AI story that involves a hardware monopoly buying a French chatbot startup.
Thesis: ASML is trading easy, high-volume DUV revenue (China) for hard, engineering-heavy EUV revenue (Hyperscalers). While revenue looks flat-to-up, the quality of that revenue is degrading—higher concentration, tougher customers (TSMC), and higher technical risk. The Mistral AI investment is a glaring signal that they feel the need to manufacture an 'AI software' narrative because the hardware monopoly alone isn't generating the multiple they want.
Verdict: AVOID — Conviction: HIGH
Catalyst: Q1 2026 guidance will reveal if the non-China EUV ramp is actually steep enough to cover the evaporation of Chinese DUV orders.
Key Risk: If Hyperscaler capex pauses even for two quarters, the 'EUV offset' thesis collapses, as there is no legacy business left to cushion the fall.
The Tell: The CFO explicitly framing the Mistral partnership as 'strategic' because 'AI could create a lot of value in our product.' A semiconductor equipment monopoly shouldn't need a minority stake in a generic LLM provider to optimize its own metrology code. It's window dressing.
Friction Level: HIGH_FRICTION — Management claims EUV growth fully offsets the China collapse. The math requires perfect execution on High-NA ramp to fill a massive structural hole.
Report not found
The report data is no longer available. Please return to the archive.