High Signal Podcasts Evidence ledger
Method
Browse
← Back to evidence

Evidence receipt / belief

Published · transcript-backed

Dylan Patel: belief

13 Mar 2026 Dwarkesh Podcast Dylan Patel — Deep dive on the 3 big bottlenecks to scaling AI compute

“There are a couple of companies and I think that could be a big disruption to the industry beyond EUV.”

— Dylan Patel

Source trail

Everything needed to verify it.

Speaker
Dylan Patel
Attribution
Verified speaker
Claim type
belief
Recorded
13 Mar 2026
Publisher
Dwarkesh Podcast

Transcript context

…How surprised would you be if in 2030 there just happened to be some total disruption where we’re not using EUV? What if we’re using something that has much better effects, is much simpler to produce, and can be produced in much bigger quantities? I’m sure as an industry insider that sounds like a totally naive question, but do you see what I’m asking? What probability should we put on something coming totally out of left field to make all of this irrelevant? Something that’s very simple and easy to scale, I assign a very, very low probability. There are a number of companies working on effectively particle accelerators or synchrotrons that generate light that’s either 13.5 nanometer, like EUV, or an even narrower wavelength, like X-ray at 7 nanometers, to then use in lithography tools. But those things are massive particle accelerators generating this light. It’s a very complicated thing to build. There are a couple of companies and I think that could be a big disruption to the industry beyond EUV. But I don’t think we’re going to magically build something new that is direct write and super simple, and can be manufactured at huge volumes, although there are some attempts to do things like this. I ask because if you think about Elon’s companies in the past, rocketry was this thing that was thought to be—and is—incredibly complicated.…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.

Search evidence