4 quotes from 1 episode on Odd Lots, each with a timestamped link to the source.
4 quotes1 episode
The short version
Stanley Lee says high GPU rental prices indicate the high cost of capital for buyers. The spot compute market operates on thin volume, allowing a tiny fraction of trades to set public benchmark rates.
Most interesting insights
Spot compute behaves like crude oil where a tiny fraction of total volume sets the public benchmark rate.
“This market is thin. It's It's like other commodities like oil, where a vast minority of the amount of compute that's procured and traded is done so in the spot market, but these are the prices that we use as our reference rates.”
Nvidia actively finances the market by injecting equity into AI labs and offering balance sheet backstops to high-yield credit lenders.
“Nvidia has been very relevant in helping to finance both sides of the equation. Nvidia has backstopped and invested in many of the neolab equity raises. They have also gone to the credit side and provided backstops to the creditors.”
High on-demand pricing reveals the actual cost of capital for buyers. Tight credit conditions drive these rising GPU rental indices.
“And this is why, you know, spot pricing and on-demand pricing and short-term pricing is remaining so high, because it's so indicative, not just of the underlying demand, but of the buyers' cost of capital…”
Compute customers want to sign multi-year capacity agreements. Private credit lenders demand steep prepayments and strict loan-to-value ratios, shutting smaller teams out of term deals.
“A lot of the buyers of spot, on-demand, and short-term pricing wish that they could buy term. They cannot…”
Stanley Lee points out that Nvidia actively finances both sides of the market, injecting equity into AI labs while offering balance sheet backstops to high-yield credit lenders.
Paul Kedrosky describes Nvidia as the private central bank of AI in 2026, using its balance sheet to create credit from nothing through delayed payment terms and neocloud equity.
How we attribute quotes. Every quote was matched against the episode transcript, so the words and the timestamp are real (we trim filler words like "um", nothing else). The name comes from our written summary of the episode, and we use it only when a separate check of the captions finds that person on the episode. YouTube gives us no voice-by-voice transcript, so open the timestamp to hear who is talking. See a wrong name? Tell us and we fix or remove it.