"Here Comes AI That Pays for You"... Meta's "Muse" Kicks Off Stablecoin Trend [Bitcoin Now]

AI Now Handles Everything from Subscription Cancellations to Payments

Currently Uses Existing Streamlined Payment Services

Stablecoins Expected to Be Introduced for Micro and High-Frequency Transactions

Meta's personal artificial intelligence (AI) agent, "Muse," has begun participating in economic activities on behalf of humans. Although it is still the early stage of agentic commerce, Muse is drawing attention for handling subscribers' cancellations and facilitating online payments, which has spotlighted AI agent-based stablecoin payments.

According to iM Securities, if micro and high-frequency transactions increase as data and external application programming interfaces (APIs) are purchased more frequently, it is expected that the potential use cases for stablecoins will also expand.

Muse, unveiled by Meta on September 8, is capable of tasks such as sending emails, booking travel, and making online purchases. More recently, it has attracted attention by adding a feature that helps users find and cancel forgotten paid subscriptions. In this process, AI substitutes the consumer in searching for cancellation methods and executing the necessary steps. This capability suggests that, in the future, such functions could save consumers time and money, and AI could further develop into an agent capable of executing economic decisions.

Currently, Muse is in its initial commercialization phase, where it performs economic activities with user authorization. For purchases made through Muse, Meta uses Stripe's streamlined payment service "Link." On participating retailers, Muse utilizes the user's registered payment method; for other businesses, it employs single-use virtual cards limited to authorized purchases. Users approve the transaction amount each time a purchase is made.

Although the official announcement does not specify direct stablecoin payments through Muse or the adoption of on-chain settlements, there is a possibility that, in the future, AI agents could use stablecoins while calling data or APIs. For example, a travel-planning agent could buy real-time fare data or use external search services before booking a flight. Agents used for corporate work could also flexibly use and pay for data, APIs, or computing resources as needed. As the scope of these activities broadens, the potential for micro and repetitive payments between services, beyond just final product purchases, is likely to increase.

In an environment where extremely small amounts are paid at high frequency, the ability to reduce transaction fees and automatically process payments and settlements according to predetermined conditions—i.e., smart contract functionality—becomes increasingly important. Stablecoins are considered a meaningful option as they can flexibly support micro, high-frequency transactions and smart contract settlements.

The role of Stripe, Muse's payment partner, is also worth noting. In addition to supporting consumer purchases via Link, Stripe provides payment technology for agents to access digital services and make payments. The "Machine Payments Protocol (MPP)," jointly announced by Stripe and Tempo in March of this year, is also part of this trend. MPP is an open standard that allows services to present usage fees, enables agents to make payments, and grants access to necessary resources. It supports stablecoins and traditional payment methods like cards, with revealed use cases such as payments based on API calls or browser usage. As such, a foundational system for agents to make payments for external services, alongside consumer purchase support, is being established.

If Muse expands its use of external paid services, there will be further opportunities to leverage the functions of Stripe, its existing partner. However, it is important to distinguish between Link's general support for stablecoins, the payment functions of MPP, and their actual implementation in Muse. For Muse to implement such features, it would require additional integrations, spending permissions, and wallet or fund management frameworks. At present, it is more appropriate to view these efforts as laying the groundwork for future expansion of payment methods, rather than confirmed feature rollouts.

Global asset management firm BlackRock also suggested a complementary development of using legacy payment networks for consumer purchases and stablecoins for machine-to-machine microtransactions, in its report "The Machine-Native Economy" released on September 23 (local time).

In BlackRock's example of travel booking, a primary agent checks the user's schedule and information via the Model Context Protocol (MCP) and delegates tasks to a travel specialist agent through the Agent-to-Agent (A2A) protocol. The secondary agent uses the x402 machine-to-machine payment protocol to purchase the required fare and room data, after which the primary agent completes the airline and hotel bookings and payment process via the Agentic Commerce Protocol (ACP). It is a structure where transactions to secure the data needed for purchase decisions take place separately before the transaction for the final travel product itself.

BlackRock anticipates that both traditional payment networks and blockchain technology will be utilized in parallel. In areas that connect consumers and existing businesses, the role of cards and similar payment methods will continue, while digital assets such as stablecoins could be used for micro, high-frequency machine-to-machine transactions. Applied to Muse, this means consumers could pay for travel bookings and agent service fees using traditional methods, while microtransactions for internal data purchases between services could use stablecoins.

Kyuwon Lee, a researcher at DS Investment & Securities, analyzed, "As personal AI agents increasingly handle product discovery, price comparison, and payments, the initiative in transactions is shifting from traditional commerce or travel platforms to agents." He added, "Because consumers no longer visit platforms directly, B2C companies now face a shift to a 'B2A (Business to Agent)' structure, where they must be chosen by agents instead of people. This has launched the 'Muse Trade' phenomenon, where there is a growing disparity between AI infrastructure beneficiaries and traditional consumer platforms."

This content was produced with the assistance of AI translation services.