In today’s newsletter, Vincent Chok from First Digital reveals the rise of “agent finance,” where artificial intelligence agents are going beyond advice to execute financial transactions, making cryptocurrencies an important financial backend for this machine-driven economy.
Then, in Ask the Expert, we asked three leading AI systems (Grok, Gemini, and Claude) two questions about the necessary steps for AI payments use cases and scalability.
Note: Answers are generated by an AI assistant and reflect the perspective of each model. They should not be considered financial or legal advice.
——Sarah Morton
Artificial Intelligence Agents in Cryptocurrency: What Advisors Need to Know
The explosion of AI agents
Artificial intelligence agents have become one of the hottest topics of the past year. A recent PwC survey of more than 300 companies found that 79% already employ AI agents in some form. This explosion reflects a broader shift: AI agents are evolving from advisory to executive roles.
AI systems initially deployed to aid chatbot services and co-pilot roles are now actively planning, deciding and acting based on predefined parameters set by humans, including financial transactions. The result was the early formation of “agency finance”. This is a new primitive in which an AI agent essentially performs financial actions within predefined rules such as limits, permissions, and goals.
Breaking down agency finance
Agency finance can be understood from three levels. The agent business layer focuses on discovery and decision-making. For example, an AI agent can search for the best hotel deals for an upcoming trip. The agent payment layer is responsible for execution, and the agent completes the transaction after approval.
Finally, asset management represents the entire stack, where agents can manage portfolios, process payments and dynamically optimize financial strategies based on real-time market trends. While this may seem like we are giving our AI agents complete autonomy, this is not the case. It is a conditional delegation where the user retains control through constraints while offloading execution.
In theory, AI agents do have a use case in finance; however, they don’t exactly fit into existing traditional financial infrastructure. Structurally, AI agents do not have direct access to global banking tracks and are designed to operate 24/7. This structural mismatch is where cryptocurrencies come into play.
Stablecoins provide AI agents with access to programmable, always-on funds, blockchain enables instant and global settlement, and crypto wallets provide permissionless access to funds. Essentially, these components form a financial layer that is more suitable for machine-driven activities. As a result, cryptocurrencies are increasingly becoming the infrastructure of autonomous systems rather than just an asset class.
Use cases for AI agents
Early implementations are already visible. Machine-to-machine payments powered by API access and data providers make merchant-to-merchant rails stronger and faster. On the consumer side, Autonomous Commerce allows users to optimize retail research and leverage agents to get the best deals on travel, subscriptions and shopping.
Meanwhile, in cryptocurrency-native environments, trading agents are widely deployed for portfolio management, return optimization, and trading strategies. On the enterprise side, supply chain management and supplier payments are easily automated through AI agents, reducing errors and resource expenditure. At this stage, most activity is still business-to-business and infrastructure-driven rather than consumer-facing.
Beyond use cases, AI agents play an integral role in driving demand for new investable categories, as well as for cryptocurrencies themselves. Since AI agents cannot run on existing infrastructure rails, there is a growing need for agent-native wallets, stablecoin payment rails, and data or computing markets.
For example, Coinbase launched x402, an open payments protocol designed to broker native transactions. This shift is particularly important for micropayments, where high transaction volumes and low value make traditional rail inefficient. For the first time, non-human users are participating in financial systems and driving activity. Artificial intelligence agents have become a new class of “users” of crypto networks.
Risks and Future Outlook
Despite the strong momentum, we are still in the early stages and there are risks and limitations. Security is a primary concern, especially when it comes to rogue or exploited agents executing unintended transactions. Issues regarding authorization, responsibilities and regulatory processing remain under review and are being actively defined. For widespread adoption, we must build user trust. This is achieved through regulatory clarity for all relevant stakeholders, so projects can be built with clarity and confidence while protecting user funds and interests.
Over the next twelve months, the technology will continue to develop and mature. Important signals include the growth of agent-driven trading volumes, the emergence of agent-native wallets and payment protocols, and deeper integration between stablecoins and AI-driven systems. Finally, regulatory clarity will largely influence the pace and scope of adoption across different industries and sectors.
In summary
Artificial intelligence agents are not a theoretical concept; They already execute transactions in a limited environment. As the trend continues, cryptocurrencies are increasingly becoming the financial backend of machine-driven economies. For now, this is an infrastructure and long-term theme play; however, this is changing as adoption rates increase. Advisors should consider this as the next wave driver of cryptocurrency utility.
– Vincent Chok, CEO and Co-Founder, First Digital
Ask the experts
This week we’re going to do something a little different. We don’t have one expert, but a group of experts—artificial intelligence experts. Below, we asked three leading AI models two questions about the current state and future of AI-powered payments. While there are common themes – especially around what is needed for future growth – there are also some clear differences. We hope you find this experiment as interesting and thought-provoking as we did.
Q1: What AI payment use cases do you see today?
GronkxAI |
Fraud detection is the most mature use case – AI can spot anomalies that rule-based systems miss. Smart payment routing dynamically selects the best rail based on cost and speed. Emerging agent payments enable autonomous AI to handle B2B financial tasks and machine-to-machine micropayments using stablecoins, and Visa, PayPal, and Mastercard are starting to validate these agents. |
|---|---|
| Gemini Google | Three “agent commerce” breakthroughs stand out: autonomous treasury and DeFi revenue 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, where AI-driven direct indexing handles partial transactions and daily tax loss harvesting at a scale that human advisors cannot manage manually. |
| Claude Human | AI payments are moving from concept to infrastructure faster than most advisors realize. The clearest use case: Programmatic billing for AI services and agents, automatically spending money on calculations or subtasks without requiring human approval for each transaction. On the wealth side, AI is enhancing rebalancing and tax-loss harvesting, while humans are still involved in the cycle of most major decisions today. |
Q2: What is needed to scale up artificial intelligence payment?
GronkxAI |
Three gaps need to be filled: standardized agent identities with cryptographic proof of authorization and clear accountability rules; governance and explainability for regulators, including anti-money laundering compliance in automated wealth flows; and high-quality, real-time data infrastructure connecting fiat and cryptocurrencies. Advisors must demonstrate that AI augments, rather than replaces, fiduciary duties. |
|---|---|
| Gemini Google | Three things must come together: verifiable agent identities authorized using DID and cryptographic signatures so that institutions can distinguish authorized AI from rogue bots; a standardized governance protocol with universal spending limits and a kill switch enforced at the protocol level; and a legal liability framework that makes it clear who is responsible when an AI performs flawed transactions or falls victim to an instant injection attack. |
| Claude Human | Three things that are really missing are: identity, accountability, and spending control. AI agents lack a standardized identity layer—and therein lies the trust risk hidden in all the excitement. There needs to be programmable guardrails built into the payment layer (such as corporate cards with hard limits and audit trails), as well as regulatory clarity for stablecoins. The technology is almost ready. What’s lagging behind is governance – and that’s an opportunity for advisors who are ahead of the curve. |