Buying Clouds with Smoke
I saw the headlines screaming this morning: Meta signs a massive 1.2 GW nuclear deal with Oklo. The stock market, predictably, went into a frenzy. Oklo shares surged. Meta gets to claim it’s “powering the future.”
But my mind ponders - are we celebrating the purchase of electricity, or the purchase of a dream?
The Math of “Future Power”
Let’s strip away the press release optimism and look at the engineering reality.
-
The Promise: 1.2 Gigawatts (1,200 Megawatts) of power by 2034.
-
The Unit: Oklo’s flagship “Aurora” powerhouse produces roughly 15 MW to 50 MW.
To hit 1.2 GW, Oklo doesn’t just need to build one reactor. They need to manufacture and deploy somewhere between 24 and 80 reactors in Southern Ohio.
-
Current Operational Reactors: 0
-
NRC Licenses Approved: 0
-
Construction Started: 0
They are promising to build an entire fleet of nuclear reactors in 8 years, using a design that the regulator (NRC) effectively sent back to the drawing board just a few years ago.
The Pre-Payment Paradox
Meta isn’t buying power that exists. They are “pre-paying” to fund the construction of the power plants. This isn’t a utility bill; it is venture capital disguised as procurement.
Big Tech is so desperate for energy to feed their AI clusters that they are now bankrolling the R&D of unproven nuclear startups. They are buying call options on physics.
Is it real?
I find the ambition admirable. We need nuclear. But there is a fine line between a “strategic roadmap” and “science fiction.”
We are celebrating a timeline (2030 start) that assumes everything goes perfectly in an industry notorious for never going perfectly. If the tide is rising, and your boat is still just a blueprint in a regulator’s filing cabinet, are you really swimming?