In today’s newsletter, First Digital’s Vincent Chok discusses the rise of “agent finance,” where AI agents are going beyond advice to execute financial transactions, making cryptocurrencies the essential financial support for this machine-driven economy.
Then, in “Ask an Expert,” we posed two questions to three leading AI systems (Grok, Gemini, and Claude) about AI payments use cases and the steps needed for scalability.
Note: Responses were generated by AI assistants and reflect the perspective of each model. They should not be construed as financial or legal advice.
– Sara Morton
AI Agents in Crypto: What Advisors Need to Know
The explosive growth of AI agents
AI agents have become one of the hottest topics in the last year. A recent PwC survey of more than 300 companies found that 79% are already adopting AI agents in some form. This explosive growth reflects a broader shift: AI agents are evolving from advisory roles to execution roles.
Initially deployed to assist with chatbot services and co-pilot roles, AI systems now actively plan, decide and act based on predefined parameters set by humans, including financial transactions. The result is the early formation of “agent finance.” This is a new primitive where AI agents essentially execute financial actions within predefined rules such as limits, permissions, and goals.
Breaking down agent finances
Agent finance can be understood in three layers. The agent trading layer focuses on discovery and decision making. For example, an AI agent can search for the best hotel deal for an upcoming trip. The agent payments layer handles execution, where the agent completes a transaction once approved.
Finally, the asset management layer represents the complete stack, where the agent can manage portfolios, manage payments and dynamically optimize financial strategies based on real-time market trends. While it may seem like we are giving full autonomy to AI agents, that is not the case. It is a conditional delegation, in which users retain control through constraints while offloading execution.
In theory, AI agents have a use case in the financial space; however, they do not fit neatly with the existing traditional financial infrastructure. Structurally, AI agents lack direct access to global banking avenues and are designed to operate 24/7. This structural imbalance is where cryptocurrencies come into play.
Stablecoins offer AI agents access to programmable, always-available money, blockchains enable instant and global settlements, and crypto wallets provide permissionless access to funds. Basically, these components form a financial layer that is best suited for machine-driven activity. Cryptocurrencies are therefore increasingly becoming the infrastructure for autonomous systems, rather than just an asset class.
AI Agent Use Cases
The first implementations are already visible. Machine-to-machine payments powered by API access and data providers have made pathways between merchants stronger and faster. In the consumer context, autonomous commerce has allowed users to optimize retail research, using agents to get the best deals for travel, subscriptions, and purchases.
Meanwhile, in crypto-native environments, trading agents are widely used for portfolio management, performance optimization, and trading strategies. On the business side, supply chain management and supplier payments have been easily automated through AI agents, reducing errors and resource waste. At this stage, most activity remains business-to-business and infrastructure-driven, rather than consumer-facing.
Beyond use cases, AI agents also play an integral role in driving new investment categories, as well as demand for cryptocurrencies themselves. As AI agents cannot operate on existing infrastructure pathways, demand is growing for native agent wallets, stablecoin payment pathways, and data or compute marketplaces.
Coinbase, for example, launched x402, an open payments protocol designed for native agent transactions. This change is particularly relevant for micropayments, where high transaction volumes and low value make traditional railways inefficient. For the first time, non-human users participate in the financial system and conduct the activity. AI agents have become a new class of “user” for crypto networks.
Risks and future prospects
Despite the momentum, we are still in the early stages and there are risks and limitations. Security is the primary concern, particularly around dishonest or exploited agents executing unwanted transactions. Issues relating to authorization, liability and regulatory treatment are still under scrutiny and are being actively defined. For widespread adoption, we must build trust among users. This is achieved through regulatory clarity for all stakeholders involved, so that projects can be built with clarity and confidence while safeguarding user funds and interests.
Over the next twelve months, this technology will continue to grow and mature. Signs that matter include growth in the volume of agent-powered transactions, the emergence of native agent wallets and payments protocols, and deeper integration between stablecoins and AI-powered systems. Finally, regulatory clarity will largely determine the pace and extent of adoption across different industries and fields.
In conclusion
AI agents are not a theoretical concept; They are already running transactions in sandboxes. As the trend develops, cryptocurrencies are increasingly emerging as the financial backstop for machine-driven economies. For now, it is a long-term infrastructure thematic work; However, that is changing with increasing adoption rates. Advisors should consider it as a driver of the next wave of crypto profits.
– Vincent Chok, CEO and co-founder of First Digital
ask an expert
This week we will do something a little different. Instead of one expert, we have a panel of experts: AI experts. Below, we posed two questions to three leading AI models, asking about the present and future of AI payments. While there were common themes, especially around what is needed for future growth, there are also some clear differences. We hope you find this experiment as fun and stimulating as we do.
Q1: What AI payments use cases are you seeing today?
Grok x AI |
Fraud detection is the most mature use case: AI detects anomalies that rules-based systems miss. Intelligent payment routing dynamically selects the best rail by cost and speed. Emerging agent payments allow autonomous AI to handle B2B treasury tasks and machine-to-machine micropayments using stablecoins, and Visa, PayPal, and Mastercard are starting to verify these agents. |
|---|---|
| Gemini Google | Three advancements of “Agentic Trading” stand out: autonomous treasury and DeFi yield management, where AI executes cross-chain exchanges in real-time; the “Inference Economy,” where agents instantly settle machine-to-machine micropayments via stablecoins; and hyper-personalized wealth implementation: AI-powered direct indexing that handles fractional transactions and daily tax-loss harvesting at a scale no human advisor could manage manually. |
| anthropic claude | AI payments are moving from concept to infrastructure faster than most advisors realize. Clearer use cases: programmatic billing for AI services and agents that autonomously spend funds on computation or subtasks without human approval per transaction. On the wealth side, AI is increasing rebalancing and tax-loss harvesting, and humans are still aware of most material decisions today. |
Q2: What will it take for AI payments to grow?
Grok x AI |
Three gaps need to be closed: standardized agent identity with cryptographic proof of authorization and clear accountability rules; governance and explainability for regulators, including AML compliance in automated wealth flows; and high-quality real-time data infrastructure that bridges fiat and crypto. Advisors must demonstrate that AI augments, rather than replaces, fiduciary duty. |
|---|---|
| Gemini Google | Three things must converge: DID-verifiable agent identity and cryptographically signed mandates so that institutions can distinguish authorized AI from unauthorized robots; standardized governance protocols with universal spending limits and kill switches applied at the protocol level; and legal liability frameworks that clarify who is responsible when an AI performs a faulty operation or is the victim of an immediate injection attack. |
| anthropic claude | Three things are really missing: identity, accountability and spending controls. AI agents lack a standardized identity layer – that’s the fiduciary risk behind all the excitement. Programmable guardrails natively integrated into the payment layer (such as a corporate card with strict limits and audit trails) are needed, along with regulatory clarity around stablecoins. The technology is practically ready. What is lagging is governance, and that is an opportunity for advisors to get ahead of it. |
