BlackRock: AI Agents Could Drive Crypto's Next Demand Wave
BlackRock's Digital Assets Research team argues in a paper titled "The Machine-Native Economy" that autonomous AI agents paying for services could become a major driver of crypto demand, with stablecoins positioned as the most suitable payment rail. The report also proposes tokenized, standardized claims on computing power as a new digital asset class and notes existing agent-payment plumbing such as Coinbase's x402 and cloud-provider integrations.

Why It Matters
If agentic AI begins making high-frequency, micro-value payments on its own, conventional banking and card rails may be ill-suited to that volume and latency, potentially shifting settlement and financing activity onto blockchains and stablecoins. That dynamic could materially increase on-chain transaction flows and create new markets tied to compute capacity.
Key Facts
- Paper title: The Machine-Native Economy
- Authors / teams: BlackRock Digital Assets Research (led by Will Su and Robert Mitchnick) with U.S. equity ETF and iShares product leads
- Stablecoin adjusted transaction volume (2025): $11 trillion
- Traditional ACH volume (2025): $93 trillion
- Stablecoin annual growth since 2020: ~80% per year
BlackRock's Digital Assets Research team published a paper arguing that agentic artificial intelligence—systems that can execute multi-step tasks and transact with external services autonomously—could drive substantial new demand for crypto infrastructure. The report positions stablecoins as a natural payment instrument for such machine-to-machine commerce because they can settle quickly, operate 24/7, and avoid the human identity and fee structures inherent to traditional bank and card rails.
The paper highlights existing technical building blocks for agent payments. Coinbase's x402 protocol, which repurposes the HTTP 402 status for payment-required requests, is cited as a way that software can pay for data feeds or API calls in the same request. Major cloud and platform providers have begun experimenting with these flows: Amazon has tied stablecoin payments into some AI cloud tools via Coinbase and Stripe, and Google extended its Agent2Agent framework to support cards, stablecoins, and real-time bank transfers with support from Coinbase and the Ethereum Foundation.
Real-world agent-originated payment volume remains small today. Blockchain analytics firm TRM Labs reviewed $52.7 million of x402 settlements and estimated that autonomous AI agents likely accounted for between 0.6% and 7.5% of that value, with much of the traffic resembling automated scripts rather than fully agentic systems. BlackRock does not argue immediate scale but contends that the foundational plumbing is being put in place ahead of broader adoption.
Beyond payments, the paper floats a new asset class: standardized, tokenized claims on computing capacity. BlackRock envisions compute packaged into tradable contracts that can be bought, sold, financed, or used as collateral and settled programmatically on-chain. The firm points to rising cloud revenue forecasts—analyst estimates put combined 2030 revenue for Amazon, Microsoft, and Google's cloud divisions near $1.1 trillion—as evidence of growing demand for compute that could be allocated via such markets. BlackRock also reports stablecoins had a circulating market capitalization above $300 billion as of September 2026 and reiterates its view that agentic AI, rather than regulation or fresh institutional entrants, could be a key driver of future crypto demand.
Keep Reading

Live updates: Bitcoin slips as higher rates and dollar pressure markets

MoonPay to acquire SEC-registered North Capital in $60 million all-stock deal

Bitcoin's $16 billion quarterly options settlement arrives with a 'call-heavy' book
