We scan new podcasts and send you the top 5 insights daily.
Evan Conrad says compute's forward curve can be backwardated or contango and often looks humped. Unsold capacity approaching delivery cannot be sold in the past, so it drops price to clear; later capacity may price lower because buyers pay for shorter lead time. He compares expiring capacity to discounted milk.
AI companies with the foresight to sign long-term, multi-year compute contracts gain a significant margin advantage. They lock in prices based on past valuations, while competitors are forced to buy capacity at much higher current market rates driven up by the increasing value of new AI models.
Evan Conrad describes the order book as a market for breaking leases. Batch inference can accept interruption in minutes; live inference customers may need months of notice and face SLA penalties. Those interruption rights price differently, similar to power markets' day-ahead and live markets. The terms are the product.
In a backwardated market, futures contracts for later delivery are cheaper than the current one. This allows investors to generate a "positive roll yield" by selling high and buying low on each contract roll. This can boost total returns significantly beyond what spot price movement alone would suggest, creating a powerful tailwind.
Don Wilson rejects the idea that futures simply let firms borrow more against the same risk. He says every contract needs a rational buyer, and heavy hedging by compute sellers should push forward prices lower, forcing investors to buy them only if they expect a return.
Previous attempts at tech futures like DRAM failed because prices only moved in one predictable direction: down. In contrast, the market for GPU compute will experience cycles of high demand and excess supply. This two-way volatility creates genuine hedging needs, making a futures market viable and necessary.
Evan Conrad estimates a B300 can make about $40/hour selling inference at full utilization, or $25/hour at 50%, while B300s rent for $5-6/hour. He says one of those numbers must change, calling it a huge arbitrage. That gap shapes willingness to pay for future capacity.
The head of AI at Hudson River Trading highlights a practical barrier to creating a financial market for compute. For serious training, the minimum "lot size" is thousands of GPUs, not a small, fungible unit. This makes it difficult to standardize a contract and create liquidity, unlike commodities with smaller, interchangeable units.
The GPU rental forward curve has shifted up and flattened, moving from backwardation toward contango. This shows providers are no longer offering deep discounts for long-term contracts, signaling their confidence that demand will remain strong and they will have opportunities to raise prices in the future.
Evan Conrad argues financializing compute before physical settlement creates a market that is easy to manipulate and carries basis risk: the hedge price may not match actual deliverable capacity. SF Compute co-designed its order book with physical constraints; Conrad says cash settlement is good only if derived from physical settlement.
The CME is creating futures contracts based on H100 and B200 GPU rental price indices. This marks the financialization of compute, allowing providers to hedge revenue and enterprises to lock in future costs, with a long-term vision for physical delivery of compute capacity.