We’ve designed and built our first AI chip: Jalapeño.
Designed from the ground up by OpenAI and brought to production with @Broadcom, Jalapeño is purpose-built for the LLM workloads powering ChatGPT, Codex, the API, and future agentic products.
Chips are foundational to the AI economy. Building our own expands our full-stack platform from products to models to infrastructure, and will help us scale intelligence, serve more people, and expand access to AI.
I mean... if i had to guess, $LPK did claim "80% of customers among major global players have selected LPKF equipment".
Kinda clear glass substrates is the next packaging shift!
From est. timelines, major players like Absolics is ramping H2 2026, Samsung Electro Mechanics 2027 + their partners.
Then there's major shifts like TSM CoPoS 2028 (seems possible $LPK upstream exposure, unconfirmed).
Just that claim of 80% is staggering and my personal expectation was markets might price it in eventually as they volume ramp (disclosure, own LPK).
Don't think there needs to be any major news, probably just getting closer to HVM timelines.
$XFAB power semis segment does not have exposure to AI.
X-FAB’s GaN platform uses depletion-mode (dMode) architecture. Datacenter power conversion has standardized on enhancement-mode (eMode) GaN because eMode is normally-off, safer, simpler gate drive circuitry, easier to design with.
Navitas, GaN Systems (now Infineon), and Power Integrations all use eMode. X-FAB’s dMode GaN is not interchangeable, it requires completely different circuit topology. No datacenter power designer is going to redesign their reference architecture around dMode GaN when the entire industry ecosystem of drivers, controllers and reference designs is built for eMode.