Invest1 publisher2 min readPublished
BlackRock's research team sizes tokenized compute against $1.1 trillion of 2030 cloud revenue
The paper proposes standardized digital claims on compute as tools for financing and settlement, and puts hyperscaler cloud revenue at roughly $1.1 trillion a year by 2030 while calling such markets largely speculative at this stage.
The Investor · Invest desk

What happened
- BlackRock's Digital Assets Research team published a white paper arguing that standardized digital contracts, essentially claims on compute resources, could be used for financing and programmable settlement.
- It names three areas where AI and programmable finance are converging: automated machine-to-machine payments, tokenized financial assets, and emerging markets for computing capacity.
- The document is research: it announces no new products and does not reference any existing tokens.
Compiled by The InvestorSomething wrong?How this is made
Why it matters
- constraint A claim only trades if buyers treat every unit as interchangeable, and the paper's own use case has agents selecting providers on price, performance and latency, so the standardization that makes a contract liquid strips out what the buyer is shopping for.
- decision Nothing here is a bid for capacity, so an operator deciding how to fund a build cannot raise against a BlackRock claim on it until the firm files something.
- contradiction One document both sizes a $1.1 trillion revenue line and calls the markets for it largely speculative, so anyone quoting it as validation is quoting the half they prefer.
- precedent BlackRock's on-chain products so far wrap instruments whose legal claim already exists, a bar a compute claim has to clear before a regulated fund can hold one.
Work the growth rate backwards. A 29% compound rate over five years is a multiple of about 3.57, so roughly $1.1 trillion in 2030 implies about $308 billion of hyperscaler cloud revenue in 2025 [5][13][14]. That line is what cloud customers pay hyperscalers in a year [5]. A market in standardized claims on capacity would be a different number: the share of capacity sold forward, times the number of times each claim changes hands before delivery. Crypto Briefing, which reported the paper, treats the $1.1 trillion as a concrete target for how large the addressable market could become, and says even a small percentage of it would be bigger than most existing crypto sectors [16]. One percent is $11 billion a year [15].
Standardization is the hard part. A claim on a GPU-hour only trades if buyers will accept any unit of it. Naming the chip, the region and the delivery window narrows the pool that will take delivery. The paper's own use case has AI agents choosing providers on price, performance and latency [4].
"Compute is emerging as a new and potentially large market for digital assets," the paper says [6]. It also calls these markets "largely speculative at this stage" [10], and says infrastructure for permissions, identity and compliance has to be built before any of it becomes real [11].
BlackRock's existing on-chain work is the BUIDL fund and tokenized money-market products, which put traditional financial instruments on chain [12]. The legal claim behind a money-market share is settled; ownership and redemption are defined. Compute has no equivalent yet, and the permissions and compliance caveat covers that gap [11].
There are a couple of other ways this goes. The paper stays a research document, cited by promoters and never turned into a product. Or a compute provider starts posting forward capacity with a delivery spec and a price, and this paper looks like early positioning on a market someone else builds. Or the payment rails arrive first, x402 and the Machine Payments Protocol among them [8], and capacity keeps changing hands on bilateral contracts with agents paying per call.
The falsifier is specific: a claim naming a chip, a location and a delivery window, quoted at a price a seller will clear. Until one of those trades, what the paper produces is a citation, carried by a firm managing north of $10 trillion [7].
What to watch
- A BlackRock filing or fund document that references claims on compute would move this from research to product.
- A compute provider publishing a standardized forward capacity contract with a chip, a location, a window and a posted price.
- Whether hyperscaler cloud revenue tracks the 29% compound rate the paper assumes from 2025.