OpenLedger core contributor Ram Kumar expects AI agents to become early users of cryptocurrency payments, arguing that autonomous software cannot independently open bank accounts or pass know-your-customer checks required by traditional financial institutions. Speaking with Kelvin Sparks at the Wyoming Blockchain Symposium 2026, Kumar said blockchain-based wallets could give agents a practical way to pay for data, computing resources and digital services.
The argument places AI agents at the center of a potentially large new category of on-chain activity: machine-to-machine payments that are too small, frequent or automated for conventional banking systems. Rather than waiting for human approval to pay an API provider, rent server capacity or acquire a data feed, an agent could use a pre-funded wallet and execute payments under limits set by its operator.
Kumar said trading could emerge as a second major crypto use case for AI agents after payments. He described a progression in which agents begin by handling research and information-gathering tasks, then gain authority to execute transactions or manage defined trading strategies.
That model would not give software an unrestricted mandate to move funds. In practice, agent-operated wallets could be limited by spending caps, approved counterparties, predefined assets or transaction rules. Those controls will likely determine whether agent payments develop beyond narrow experiments, particularly where agents interact with decentralized financial protocols that execute transactions automatically.
Micro-payments are a natural test case
A May 2026 report from Keyrock said software programs completed 176 million on-chain transfers between May 2025 and April 2026, moving about $73 million in total value. The report put the average value of those automated transfers at 31 cents, illustrating the type of payment where blockchain rails can compete with traditional systems.
A payment worth only a few cents is difficult to process economically through card networks and many bank-based payment arrangements, where fixed charges can consume much of the transfer’s value. Public blockchain networks can support smaller transfers, although the actual cost varies sharply by network and can rise during periods of congestion.
The Keyrock figures should not be read as evidence that autonomous agents have already become a dominant source of blockchain volume. The roughly $73 million moved over the reported 12-month period remains small beside activity in major crypto markets. Yet the 176 million-transfer count points to a different pattern from conventional trading: automated systems can generate large numbers of low-value actions when a network makes such transfers inexpensive enough.
Kumar’s case is built around this operational advantage. An AI system seeking live market data, storage, computing time or access to a specialized model may need to make repeated, modest payments across several service providers. Crypto wallets and programmable smart contracts would allow those transfers to be made without requiring each service to establish a separate billing relationship with the human or company operating the agent.
Trading follows research, Kumar says
Kumar said agents may move from analyzing markets to trading them, a transition already familiar in traditional finance, where algorithmic systems execute trades under human-designed rules. Crypto markets offer additional pathways because agents can directly interact with decentralized exchanges, lending protocols and liquidity pools through wallets.
The technical possibility does not remove the risks. An agent that can execute swaps or rebalance holdings needs reliable instructions, secure access to private keys and protections against manipulated market data. Errors in an automated trading system can spread quickly when transactions settle on-chain and interact with thin liquidity.
Machine-driven order flow could also make short-term activity harder to interpret in smaller tokens. Automated strategies can react to price changes, social-media signals, liquidity movements or on-chain events faster than manual traders. Their presence alone does not guarantee higher volatility, but poorly designed strategies or crowded trading rules can amplify sharp moves in markets with limited depth.
Kumar did not present agent trading as OpenLedger’s immediate product focus. His comments instead described a sequence in which payment functionality creates the economic infrastructure agents would need before they begin taking on more complex financial tasks.
OpenLedger targets consumer model tools
OpenLedger is developing blockchain infrastructure for AI systems, with an emphasis on recording data provenance and distributing rewards to contributors. The company raised an $8 million seed round in 2024 led by Polychain Capital and Borderless Capital, then launched OPEN Mainnet in November.
According to Kumar, OpenLedger’s current platform allows users to fine-tune AI models and access models made available through its system. The company plans to add privacy options and broaden its focus from business users toward consumers over the next several years.
The consumer strategy centers on enabling people to create customized models without writing code, Kumar said. That would extend fine-tuning tools now aimed more directly at developers and enterprises. Fine-tuning adapts an existing AI model to a specific task or data set, such as a company’s internal documents, a particular writing style or a specialized support workflow.
OpenLedger’s blockchain component is intended to serve as a coordination layer for those AI activities, Kumar said. In that role, the network would record where data originated, help determine who contributed to a model or dataset, and support reward distribution among participants.
That approach addresses a growing dispute around AI development: who should receive credit or compensation when models are trained, adapted and deployed using work from multiple contributors. Blockchain records cannot by themselves resolve ownership disputes or establish whether a dataset was collected with valid permissions. They can, though, create a shared record of contributions and payments if participants choose to use the system.
Wallets would turn agents into service customers
Kumar’s vision depends on agents operating with wallets funded by the people or organizations that deploy them. Such wallets could allow an agent to pay for its own inputs while leaving ultimate control with a human owner, business or protocol.
The next test is whether those payments solve a real problem better than existing API billing, cloud credits and subscription systems. For services that require countless low-value transactions across open networks, crypto-based settlement offers a clear fit. For recurring services with established commercial relationships, conventional billing may remain simpler.
OpenLedger is betting that AI’s move from answering questions to taking actions will create more demand for the first category. If agents begin purchasing data and compute directly, the most useful crypto infrastructure may be less about speculative trading and more about inexpensive wallets, transaction controls, transparent data records and payment systems designed for software.
Explore how automation transforms trading—discover AI copy trading for agent-driven crypto strategies and smarter, hands-free market participation.
Disclaimer: The content on this page is provided for general informational purposes only and does not represent the views or financial advice of Toobit. We make no guarantees regarding the accuracy or completeness of this information and shall not be held liable for any errors, omissions, or outcomes resulting from its use. Investing in digital assets involves risk; users should independently evaluate their financial situation and the risks involved. For further details, please consult our Terms of Service and Risk Disclosure.
