AI software and the hardware that enables it have been extremely popular investments this year. But there are still limiting factors in the sector, including a lack of IT resources to power so many new start-ups. Investors don’t want to fund companies that haven’t signed a compute deal, and compute providers don’t want to sign deals for startups that haven’t yet secured funding. Now Magnetar, a hedge fund that launched its first-ever venture capital fund earlier this year, is trying to solve this “chicken and egg” problem by offering compute in exchange for equity. Magnetar was an early investor in the AI space, partnering with Coreweave and recently helping the hyperscaler raise $7.5 billion. In this episode, we talk with Jim Prusko, partner and senior portfolio manager on Magnetar’s alternative credit and fixed income team, about why the hedge fund is getting into venture capital and some of the new ways in which it deploys its money in this area. This transcript has been lightly edited for clarity.
Pod key information:
How a hedge fund got into AI – 3:05
Why investing in AI is different from previous tech booms – 6:25
A History of Asset-Based Financing – 8:46
The risks of GPU-backed loans – 11:22
The role of energy for AI startups – 14:40
Everyone is both a partner and a competitor – 19:26
How many businesses need computing to grow – 21:20
Can anyone challenge the big incumbents – 23:47
Debt vs. Equity in AI Financing – 30:02
Magnetar Presentation to Investors on AI – 34:10
The role of CoreWeave in Magnetar’s approach – 38:38
How big is the AI investment market – 44:43