Morph’s Agentic Economy Report Predicts AI Agents Will Drive $500 Billion in Commerce by 2028

Morph’s new The Agentic Economy report argues that artificial intelligence is moving beyond content generation and into a far more consequential role: becoming an active participant in the global economy. According to the report, AI agents are rapidly evolving from assistants that provide recommendations into autonomous systems capable of discovering products, negotiating purchases, authorizing payments, and settling transactions without direct human involvement. The report, published by Morph and powered by the Bitget ecosystem, outlines how new payment protocols, stablecoins, and agent-specific commerce infrastructure are converging to create what it describes as the next major phase of digital commerce.

The central premise is straightforward: the internet spent decades optimizing information exchange, but the next wave will optimize economic exchange between humans, businesses, and increasingly, machines. According to Morph, the infrastructure required for that transition is no longer theoretical. It is already being deployed by some of the world’s largest payment networks, technology companies, and commerce platforms.

The Agent Is Becoming the Buyer

One of the report’s most striking observations is that AI agents are no longer limited to answering questions or generating text. They are beginning to act as economic actors.

Morph points to several developments that occurred over the past year. Visa reported completing hundreds of secure agent-initiated transactions across its ecosystem. Mastercard expanded its Agent Pay capabilities across U.S. cardholders. Google launched its Agent Payments Protocol with more than 60 partners and integrated native stablecoin settlement functionality through x402. Stripe and OpenAI introduced the Agentic Commerce Protocol, while Shopify began rolling out agent-enabled checkout experiences across its merchant network. More than one million Shopify merchants are reportedly queued for agentic checkout capabilities.

The report argues that these developments collectively signal a shift from experimental demonstrations toward production-grade commerce infrastructure.

A Trillion-Dollar Opportunity Takes Shape

The numbers cited throughout the report illustrate why businesses are paying attention.

Salesforce reported that AI agents directly influenced $67 billion in global spending during Cyber Week 2025, representing roughly one-fifth of total spending during the period. Adobe recorded a 693% year-over-year increase in generative AI-driven retail traffic during the 2025 holiday season. Meanwhile, consulting firm McKinsey estimates that agentic commerce could represent approximately $1 trillion in U.S. retail revenue and between $3 trillion and $5 trillion globally by 2030.

Morph argues that these figures suggest agent-driven commerce is no longer a niche category. Instead, it is becoming a measurable and increasingly important component of digital retail.

Building the Agentic Payment Stack

A significant portion of the report focuses on the infrastructure required for AI agents to transact safely.

Morph breaks the emerging ecosystem into four layers: identity, mandate, checkout, and settlement. Identity systems determine whether an agent can be trusted and held accountable. Mandate systems verify that a human has authorized a specific transaction. Checkout protocols allow agents to negotiate pricing, shipping, taxes, and returns. Settlement infrastructure ultimately moves the money.

The report highlights several standards gaining traction across the industry. These include Anthropic’s Model Context Protocol (MCP), Google’s Agent Payments Protocol (AP2), the Stripe and OpenAI-backed Agentic Commerce Protocol (ACP), Shopify’s Universal Commerce Protocol (UCP), Visa’s Trusted Agent Protocol, Ethereum’s ERC-8004 trustless agent framework, and x402, a protocol designed for machine-to-machine micropayments.

Morph’s thesis is that these standards are not competing products but complementary layers of a larger technology stack that will enable autonomous commercial interactions.

Stablecoins Move to the Center

The report repeatedly emphasizes stablecoins as a foundational technology for machine-driven commerce.

Morph argues that stablecoins spent the last several years establishing themselves as infrastructure for treasury management and business-to-business payments. The next phase, according to the report, is becoming the native payment mechanism for AI agents.

One reason is economic efficiency. The report notes that the average x402 payment currently sits at approximately $0.20. Such transaction sizes are difficult to support economically through traditional card networks because of fixed processing costs. Stablecoin-based settlement, by contrast, enables sub-dollar and even sub-cent transactions between machines.

The report predicts that most agent-initiated payments, by transaction volume, will eventually settle outside traditional card rails, particularly for machine-to-machine interactions and automated service purchases.

The Market Is Growing, But Still Early

Despite the excitement, Morph also highlights the difference between infrastructure adoption and actual economic activity.

Bloomberg recently reported approximately $24 million in x402 payment volume over a 30-day period, but analysis from Andreessen Horowitz suggested that adjusted, trade-filtered activity was closer to $1.6 million during the same timeframe. The report notes that much of today’s activity still consists of testing, experimentation, and synthetic transactions rather than mature commercial demand.

What appears more significant, according to Morph, is the speed of adoption. MCP software development kit installations reportedly reached 97 million monthly installs by March 2026, while AP2 launched with more than 60 ecosystem partners. These adoption metrics suggest infrastructure is spreading faster than actual transaction volume.

Ten Predictions for the Next Three Years

The report includes ten forecasts that it expects will be resolved by the end of 2028.

Among the most notable is the prediction that agent-influenced commerce will exceed $500 billion in global gross merchandise volume by 2028. Morph also predicts that AI agents will overtake humans in commercial stablecoin payment activity, that more than one-quarter of U.S. product-discovery searches will begin in AI chat interfaces rather than traditional search engines, and that at least one major retailer will reorganize its e-commerce operations around agent readiness.

Other forecasts are more disruptive. The report suggests that autonomous agents could reduce pricing power in highly comparable product categories, driving real price declines exceeding 10% as agents comparison-shop across merchants with near-perfect efficiency. It also predicts that a Fortune 100 company will eventually attribute a major cybersecurity breach to an AI agent, forcing enterprises to rethink governance, authorization, and accountability frameworks.

Perhaps the most ambitious prediction is that one in ten U.S. households will regularly allow AI agents to complete purchases on their behalf by 2028.

How AI Agents Are Becoming Economic Actors

Whether all of Morph’s predictions prove accurate remains to be seen, but the report captures an important shift in how the technology industry is thinking about AI. For the past several years, the conversation has largely centered on productivity, content generation, and knowledge work. The Agentic Economy argues that the next chapter will be defined by economic agency: AI systems that not only generate information but also spend money, manage transactions, and participate directly in commerce.

The infrastructure supporting that transition is already emerging. Payment networks, cloud providers, commerce platforms, and blockchain ecosystems are introducing standards designed specifically for machine-to-machine transactions. At the same time, AI-powered shopping experiences are moving from experiments into production environments, giving agents the ability to compare products, negotiate purchases, and complete transactions on behalf of users.

If the trends highlighted throughout Morph’s report continue to accelerate, businesses may soon need to optimize not only for human customers, but also for AI agents acting on their behalf. In that scenario, success may depend less on brand recognition and more on whether products, pricing, and payment systems can be discovered, evaluated, and transacted upon by machines operating at internet scale. The report’s broader message is that agentic commerce is no longer a distant concept. It is becoming a measurable market, supported by rapidly evolving infrastructure and growing adoption across payments, retail, and enterprise technology.