What are the key takeaways from this Odd Lots episode?

Key insights include: **GPU Futures & Capital Costs:** Neoclouds selling compute to lower-credit customers face elevated borrowing costs, evidenced by CoreWeave bonds yielding 15%. Once GPU futures exist, a neocloud can buy hardware, deploy it, sell a strip of futures contracts, and simultaneously offer on-demand compute — structurally reducing cost of capital through a basis trade similar to commodity hedging.; **Index Robustness & IOSCO Compliance:** Not all GPU price indices are equal. Silicon Data's index focuses on on-demand neocloud pricing, where transparency is high, and is built to meet IOSCO 2013 benchmark standards. Traders evaluating GPU futures should verify whether the underlying index is IOSCO compliant before assuming it accurately reflects market prices.; **The "Crunch Spread" Framework:** Wilson introduces a new derivative concept: renting GPUs at on-demand prices, loading an open-source model, and selling tokens through platforms like OpenRouter creates a margin analogous to an oil refinery's crack spread or a soybean crusher's crush spread. DRW calls this the crunch spread — a potentially tradeable futures relationship between GPU costs and token revenue.

What did Don Wilson discuss on Odd Lots?

DRW founder Don Wilson explains why GPU compute futures markets need to exist, how they would lower capital costs for neoclouds like CoreWeave, what regulatory hurdles remain before a November 9th listing deadline, and why token futures and a "crunch spread" derivative could emerge next. Key topics include: **GPU Futures & Capital Costs:** Neoclouds selling compute to lower-credit customers face elevated borrowing costs, evidenced by CoreWeave bonds yielding 15%. Once GPU futures exist, a neocloud can buy hardware, deploy it, sell a strip of futures contracts, and simultaneously offer on-demand compute — structurally reducing cost of capital through a basis trade similar to commodity hedging.; **Index Robustness & IOSCO Compliance:** Not all GPU price indices are equal. Silicon Data's index focuses on on-demand neocloud pricing, where transparency is high, and is built to meet IOSCO 2013 benchmark standards. Traders evaluating GPU futures should verify whether the underlying index is IOSCO compliant before assuming it accurately reflects market prices..

How long is this episode of Odd Lots?

This episode is 26 minutes long. SignalCast provides an AI-generated summary so you can get the key insights in about 3 minutes.

Odd Lots

Don Wilson on the GPU Market's Trillion Dollar Question

October 19, 2026

26 min episode · 2 min read

Don Wilson

Episode

26 min

Read time

2 min

Topics

Productivity, Relationships, Investing

AI-Generated Summary

Published Oct 19, 2026

Key Takeaways

✓GPU Futures & Capital Costs: Neoclouds selling compute to lower-credit customers face elevated borrowing costs, evidenced by CoreWeave bonds yielding 15%. Once GPU futures exist, a neocloud can buy hardware, deploy it, sell a strip of futures contracts, and simultaneously offer on-demand compute — structurally reducing cost of capital through a basis trade similar to commodity hedging.
✓Index Robustness & IOSCO Compliance: Not all GPU price indices are equal. Silicon Data's index focuses on on-demand neocloud pricing, where transparency is high, and is built to meet IOSCO 2013 benchmark standards. Traders evaluating GPU futures should verify whether the underlying index is IOSCO compliant before assuming it accurately reflects market prices.
✓The "Crunch Spread" Framework: Wilson introduces a new derivative concept: renting GPUs at on-demand prices, loading an open-source model, and selling tokens through platforms like OpenRouter creates a margin analogous to an oil refinery's crack spread or a soybean crusher's crush spread. DRW calls this the crunch spread — a potentially tradeable futures relationship between GPU costs and token revenue.
✓GPU Price Volatility Is Bidirectional: H100 on-demand prices declined steadily before reversing sharply upward around Thanksgiving when Anthropic released a major model upgrade, driving inference demand. More recently, B200 and B300 prices surged while Silicon Data's token price index fell simultaneously — demonstrating that GPU compute prices can spike or collapse rapidly, making hedging instruments operationally necessary.
✓Perpetual Contracts as Democratization Tools: Perpetual futures serve two concrete use cases: eliminating roll slippage for commodity ETFs by tying them to a continuous index rather than monthly contracts, and passing through hard-to-borrow stock lending rates directly to retail holders who currently receive none of that yield from brokers. Leverage and auto-deleveraging are structural choices, not inherent properties of perps.

What It Covers

DRW founder Don Wilson explains why GPU compute futures markets need to exist, how they would lower capital costs for neoclouds like CoreWeave, what regulatory hurdles remain before a November 9th listing deadline, and why token futures and a "crunch spread" derivative could emerge next.

Key Questions Answered

•GPU Futures & Capital Costs: Neoclouds selling compute to lower-credit customers face elevated borrowing costs, evidenced by CoreWeave bonds yielding 15%. Once GPU futures exist, a neocloud can buy hardware, deploy it, sell a strip of futures contracts, and simultaneously offer on-demand compute — structurally reducing cost of capital through a basis trade similar to commodity hedging.
•Index Robustness & IOSCO Compliance: Not all GPU price indices are equal. Silicon Data's index focuses on on-demand neocloud pricing, where transparency is high, and is built to meet IOSCO 2013 benchmark standards. Traders evaluating GPU futures should verify whether the underlying index is IOSCO compliant before assuming it accurately reflects market prices.
•The "Crunch Spread" Framework: Wilson introduces a new derivative concept: renting GPUs at on-demand prices, loading an open-source model, and selling tokens through platforms like OpenRouter creates a margin analogous to an oil refinery's crack spread or a soybean crusher's crush spread. DRW calls this the crunch spread — a potentially tradeable futures relationship between GPU costs and token revenue.
•GPU Price Volatility Is Bidirectional: H100 on-demand prices declined steadily before reversing sharply upward around Thanksgiving when Anthropic released a major model upgrade, driving inference demand. More recently, B200 and B300 prices surged while Silicon Data's token price index fell simultaneously — demonstrating that GPU compute prices can spike or collapse rapidly, making hedging instruments operationally necessary.
•Perpetual Contracts as Democratization Tools: Perpetual futures serve two concrete use cases: eliminating roll slippage for commodity ETFs by tying them to a continuous index rather than monthly contracts, and passing through hard-to-borrow stock lending rates directly to retail holders who currently receive none of that yield from brokers. Leverage and auto-deleveraging are structural choices, not inherent properties of perps.

Notable Moment

Wilson reveals that the CFTC comment period for GPU futures expires November 9th — four days after a major U.S. election — and declines to speculate directly on the connection, letting the timing speak for itself in a moment that drew audible audience reaction.

Know someone who'd find this useful?

Episode Transcript

00:00:02 Speaker 1: Bloomberg Audio Studios. 00:00:04 Speaker 2: Podcasts. 00:00:05 Speaker 3: Radio. 00:00:06 Speaker 4: News. 00:00:18 Speaker 1: Hello and welcome to another episode of the Odd Lodge Podcast. I'm Joe Weisenthal. 00:00:22 Speaker 4: And I'm Tracy Allaway. 00:00:24 Speaker 1: Tracy, we're here in Chicago, second year in a row that we've done a live show here in a Great city for us. 00:00:30 Speaker 3: That's right. 00:00:30 Speaker 4: We're going to be talking to someone who we spoke with last year, but a lot has changed in 12 months. This is someone who is trying to build a market for compute, right? 00:00:40 Speaker 3: Right. 00:00:41 Speaker 1: So we're going to be speaking with Don Wilson. He is the founder and CEO of DRW. He is really at the forefront of this effort to build futures markets for things like GPUs and so forth. So check out the conversation. We talked to you, Don, basically exactly a year ago, last time we were here. We talked about things like GPU futures. But in like AI time, it might as well have been 10 years ago because so much has changed. We want to talk GPU market structure, all that stuff. 00:01:08 Speaker 2: The last couple of days. 00:01:09 Speaker 1: Have actually been kind of interesting in the AI trade world. Everyone's watching CoreWeave. Everyone's watching CoreWeave bonds. There was an anthropic investor meeting before the IPO. 00:01:19 Speaker 3: A little jitters. 00:01:21 Speaker 1: Maybe their revenue only grew up, whatever, 50% versus 51%. I don't know the exact numbers. Give us your read right now. Is there a little slowdown happening in AI? Are we shifting into a lower gear? Is there anything in the trajectory that's changed? 00:01:38 Speaker 2: Any new thing that's growing this fast is not going to grow at a constant rate. The first derivative isn't going to be constant. The second derivative isn't going to be constant. So to me, this is just kind of expected volatility. I don't know, maybe people went on vacation in August so they didn't consume as many anthropic tokens. 00:02:03 Speaker 1: But you're not reading anything more into it than that. 00:02:06 Speaker 3: I'm not. OK. 00:02:08 Speaker 4: Wait, so with something like CoreWeave, a big part of what you're doing with Compute Exchange and silicon data is this idea that if you create a market for compute, then cloud companies can hedge their future revenue streams and the cost of capital can come down. In a week like this, when you see the yield on a CoreWeave bond at 15%, what does that tell you? Is the suggestion that, well, if compute futures were up and running, all of CoreWeave's problems would be solved? 00:02:37 Speaker 2: I do think that it's highlighting the massive need for risk management products. Think about if you're a neocloud and you sign contracts …

Get the full transcript (4,452 words) + summary by email — free

One-time email with the complete transcript and AI summary of this episode. No account needed.

One email, no spam. We’ll also show you what SignalCast does.

Browse all Odd Lots transcripts →