BlackRock, the world’s largest money manager, says artificial intelligence and digital assets could become complementary foundations of an economy in which software makes purchases, moves money and obtains its own computing resources.
In its report, The Machine-Native Economy, publicized Sept. 22, authors including Will Su, head of digital assets research, argue that AI and blockchain, which have largely developed along parallel tracks, are now aligned because the two technologies are both "machine-native."
"Broad AI adoption may represent an underappreciated source of demand, utility and application growth across the digital asset economy," the authors wrote.
BlackRock, which managed $15.3 trillion at June 30, gives institutional weight to a core thesis of decentralized AI: that programmable networks could allow AI systems to exchange services and resources.
As AI agents become more capable and as their real-world applications expand, they increasingly demand payment and asset infrastructure designed natively for machine-speed commerce. Crypto-native blockchain rails are particularly well suited to high-frequency, sub-cent, machine-to-machine (M2M) transactions that take place around-the-clock, including API calls, on-demand data, and consumptionbased compute. – BlackRock
Agents buying services and computing power
Agents, which are AI-equipped programs that perform tasks with limited human intervention, could use stablecoins to make frequent payments worth fractions of a cent. Stablecoins are digital tokens designed to track currencies such as the dollar.
BlackRock illustrates this with a hypothetical travel booking: An agent receives a $2,500 budget, pays for airfare and hotel data, then books through existing merchant checkout systems.
Its examples include:
Coinbase’s x402: The cryptocurrency exchange’s open payment protocol lets software pay for online resources.
Stripe and OpenAI’s Agentic Commerce Protocol: The payments company and AI developer connect agents to businesses’ existing checkout infrastructure.
Ethereum and Circle’s Arc: The public blockchain and stablecoin issuer’s network illustrate different ways to settle payments and charge fees.
BlackRock also sees potential for AI to coordinate access to the computing power needed to train and run AI systems. Agents could compare providers and purchase capacity based on price, performance and location. Standardized capacity contracts could become tradable assets or collateral for financing.
Where BlackRock qualifies the thesis
The report acknowledges that agent payments remain nascent and compute-market liquidity is limited. (A separate story published on DeAI News on Wednesday shows the piddling level of transactions using the x402 protocol, which allows agents to make purchases over blockchain networks.)
Different chips, electricity costs and delivery arrangements complicate standardized contracts that might otherwise form the basis for AI commodity markets, although the authors consider these problems ultimately solvable.
BlackRock, which offers bitcoin and ether investment products, expects adapted traditional payment systems to remain important when agents transact with businesses and consumers. Tokenized financial assets still depend on legal frameworks, identity checks and records maintained outside blockchains.
Nor does increased blockchain use guarantee benefits for every cryptocurrency: value capture depends on network fees and incentive structures. Research suggesting AI favors digital money reflects simulated responses, the authors caution, rather than observed agent behavior.
The DeAI angle across five categories
The BlackRock report connects to several DeAI News coverage categories. The following is our editorial assessment of those connections and their limits, rather than a set of predictions from BlackRock.
| Category | The DeAI angle | Limits and open questions |
|---|---|---|
| Infrastructure & Compute | Agents shopping across computing providers and trading capacity claims could create customers and financing options for independent suppliers. | Hardware differences, regional costs and delivery standards complicate these markets. The report does not establish that decentralized providers will outperform large clouds. |
| Agentic Finance | Stablecoins and programmable assets could let agents pay for data, services and compute across independently operated systems. | Activity remains nascent. Traditional payments retain important roles, and greater blockchain usage does not automatically benefit every cryptocurrency. |
| DeAI Industry | The world’s largest money manager is giving institutional weight to AI-crypto convergence as a source of economic infrastructure and demand. | The paper is neither a commitment to finance particular projects nor evidence that their business models work. |
| Governance & Access | Standardized assets and interfaces could reduce custom integrations and help agents transact across providers. | Identity checks and legal rights still depend on institutions outside blockchains. Programmability alone does not establish open access or decentralized control. |
| Consumer | Agents could research and book travel within an approved budget, including purchasing information needed to make decisions. | The scenario is illustrative. The report does not demonstrate consumer savings or fully resolve accountability for bad purchases. |
HOW AI WAS USED IN THE PRODUCTION OF THIS PIECE: I generated the first draft of this story on ChatGPT with a custom skill and prompt, and then used our Distro Publisher MCP connector to file the piece directly to DistroVerse, our web app. I used Claude Cowork in combination with a custom skill to develop the table of DeAI angles, and used Distro Publisher MCP to insert it directly into the draft as an HTML block. I copyedited the piece with ChatGPT using the Distro Publisher MCP in combination with a custom skill.