BlackRock says the growing use of artificial intelligence could create new demand for digital assets. The stated reason was that AI systems have become more capable of interacting with money and computing resources.
In its new research paper,The Machine-Native Economy, BlackRock’s Digital Assets Research team argues that AI and blockchain are moving from separate technology trends toward a shared economic system. The paper describes AI as “machine-native intelligence” and digital assets as “machine-native money.”
BlackRock pointed out three areas where the connection could grow. This includes machine-to-machine payments, tokenized assets, and markets for computing capacity. It said stablecoins could become useful tools as AI agents begin carrying out economic tasks with less human involvement.
Our latest research paper explores the growing connection between AI and digital assets and explains why broad AI adoption may drive new demand, utility and applications across the digital asset economy. https://t.co/z5T88Orblepic.twitter.com/GMxQqTVmUN
The response from crypto and AI projects focused more on what happens when agents actually gain control over money. NeoSoul, which describes itself as building infrastructure for the agent economy, said BlackRock’s discussion of machine-native settlement was “spot on.” But it argued that payment rails are only the starting point.
read the paper and your ap2 mandates and machine-native settlement logic are actually spot on
but getting the rails right is just the tutorial level
for trading agents the real meta is earned economic authority
going from static auth to dynamic capital permissions backed by…
The company said the bigger issue was giving an agent greater access to capital when it proves it can use money effectively, and reducing that access when its performance deteriorates.This raises a question of whether an agent’s spending limits should change automatically based on its on-chain record.
That is a different opportunity from simply putting AI assets on a blockchain. Peter Schroeder, who leads global marketing for Circle, focused on another part of the BlackRock paper. He pointed to the firm’s argument that rising stablecoin activity could increase demand for blockspace and network services.Schroeder’s reaction links AI payments to demand for the networks processing them.
“As stablecoin activity scales, greater demand for blockspace and network services could support usage-related demand and potential value capture for these assets, subject to each network’s fee, staking, and gas-sponsorship design. Arc also… https://t.co/pRQpNlnlHA
Bitwise President Teddy Fusaro took a shorter view,asking why anyone would remain bearish on crypto while BlackRock itself is highlighting the connection between AI and digital assets.
Meanwhile, AiTraceRoot focused on the data layer. In itsresponse, the project argued that AI makes information usable by machines while blockchain makes assets and transactions machine-readable. That combination could give agents a way to analyse on-chain activity and then act on it.
Activity outside the paper supports BlackRock’s argument. The firm says protocols such as x402, Agentic Commerce Protocol and Google’s Agent Payments Protocol are being built to let agents access services, communicate and make payments. It also points to stablecoins as a natural fit for frequent machine transactions.
Visa and Artemis have already tracked live activity on x402. Their research found roughly $15 million in volume across 109.6 million transactions since x402 launched in May 2025. However, most payments were worth only a fraction of a cent. Visa says that type of transaction is difficult for traditional card networks because fixed fees can make tiny payments uneconomic.
BlackRock’s market view adds another reason to watch this trend. In a recentMarket Take, the firm said the AI buildout could push annual US financing needs close to $8 trillion by 2030. That means AI will need huge amounts of capital to build computing infrastructure and other services. Within the crypto sector, the opportunity could extend beyond payments to the systems used to finance, settle, and track those resources.
But trust remains a major problem. A September Visa survey found that only 23% of US consumers trusted generative AI to handle payments for them. That difference proves NeoSoul’s point. Agents will need more than wallets. They will need clear permissions, spending limits, identity controls, and records that show how they used capital.
The next test is whether autonomous agents can move beyond making recommendations and begin managing meaningful economic activity. If they can, the bigger crypto opportunity may not be AI tokens, as many people believe. It could be the rails that let software hold money, earn it, spend it, buy compute, and prove what it did with that money.
Abraham Okoh is a crypto and Web3 writer covering DeFi, digital assets, and the protocols shaping onchain finance. He writes news and analysis for DeFi Planet, and his work has previously appeared on DeFiLlama, DeFi Rate, and CryptoNewsBytes.
