Hedge fund manager Gavin Baker says investors have been too quick to judge Big Tech's AI spending

Hedge money head Gavin Baker, who famously invested early successful SpaceX, says investors person been excessively speedy to judge really overmuch Big Tech is spending connected AI infrastructure.

Baker, speaking connected the "Invest Like The Best" podcast, said improving rate flow, continued request for AI, and underpriced compute contracts are each reasons investors should interest little astir Big Tech's eye-popping spending.

"You're going to spot a batch of acceleration that's going to reply these ROI questions," Baker said, referring to what he says is simply a caller summation successful the magnitude of rate these AI businesses are collecting. "You've started to spot that this quarter."

In July, investors rushed to waste representation and spot stocks. They were mostly worried that companies for illustration Meta were spending excessively much connected AI infrastructure without a clear way to generating returns connected those investments.

Meta's free rate flow plunged 91% twelvemonth complete year, to $784 cardinal successful the 2nd 4th of 2026. The company, however, posted a 25% year-over-year emergence successful nett rate from its operating activities.

"That's a worldly acceleration astatine this scale," Baker said.

Baker besides said that persistently precocious GPU prices show that request for AI computing remains strong.

"I don't deliberation anyone successful '24 aliases '25 thought that the prices of aged GPUs would still beryllium going vertical successful 2026," he said.

Many unreality providers, Baker said, person yet to afloat monetize the AI infrastructure they've already built because overmuch of their compute capacity is tied to older, lower-priced contracts they signed erstwhile GPUs were cheaper.

"Essentially, you person the contracted guidelines of installed compute trading astatine a monolithic discount to the existent spot market," he said.