Evidence receipt / observation
Published · transcript-backedDylan Patel: observation
13 Mar 2026 Dwarkesh Podcast Dylan Patel — Deep dive on the 3 big bottlenecks to scaling AI compute
“The problem is that getting the heat out of that dense area means you have to move away from standard air and liquid cooling to more exotic forms of liquid cooling, or even immersion, to get to higher power densities.”
Source trail
Everything needed to verify it.
- Speaker
- Dylan Patel
- Attribution
- Verified speaker
- Claim type
- observation
- Recorded
- 13 Mar 2026
- Publisher
- Dwarkesh Podcast
Transcript context
…It would be better for space because more watts per millimeter means the chip runs hotter. I guess this is a question of computer chip engineering, but it cools to the fourth power by the Stefan-Boltzmann law. If you can run a very hot chip, it allows a lot of— No, you can’t run it hotter. You can only run it denser. The problem is that getting the heat out of that dense area means you have to move away from standard air and liquid cooling to more exotic forms of liquid cooling, or even immersion, to get to higher power densities. That’s more difficult in space than it is on Earth. Maybe it’s worth explaining at this point what exactly a scale-up is and what it looks like for Nvidia versus Trainium versus TPUs.…
Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.