Meta and OpenAI have released agentic products with expansive new capabilities. They’re running into a wall anyway. Consumers don’t trust them. Merchants won’t grant them access to online marketplaces. What should be a transformative shift in commerce is stalling before it starts.
According to a new WP Intelligence report from Lead AI & Tech Analyst Dan Gallagher, the barriers aren’t primarily technological. They’re structural. Agentic commerce—where AI agents autonomously handle transactions on behalf of users—requires a fundamental shift in how buyers, sellers, and platforms interact. That shift isn’t happening, and the reasons reveal how regulation and market power can slow innovation.
The Trust Problem
Consumers remain skeptical. Giving an AI agent unfettered access to one’s wallet, payment information, and purchasing history means surrendering significant control. People have legitimate questions: Will the agent make sound decisions? Could it be hacked? What happens if something goes wrong? These aren’t irrational fears. They help explain why most people haven’t adopted the technology at scale, despite its convenience.
Merchants have different but equally significant concerns. Letting agentic systems into their online marketplaces means giving up visibility and control over customer interactions. They can’t be certain how these systems will represent their products, negotiate prices, or handle disputes. From a business owner’s perspective, bringing in an intermediary—especially one powered by a competitor’s AI—creates risk without obvious compensation.
Meta and OpenAI built the tools. They’ve demonstrated technical feasibility. But they haven’t solved the human side of adoption, and that’s where commerce actually happens.
Regulatory and Competitive Headwinds
Rules for complex transactions add to the problem. Financial regulators built frameworks around human accountability and clear lines of liability. When a human financial advisor makes a mistake, responsibility is clear. When an AI agent does, it’s murkier. Who’s liable? The user? The platform? The developer? That ambiguity has slowed deployment.
Competition in tech makes matters worse. The companies building agentic products—Meta and OpenAI—also run platforms themselves. Merchants worry, with reason, that granting these companies’ agents access to their storefronts gives those companies unfair advantages. Through agentic interactions, they can observe competitor pricing, customer behavior, and market trends. That’s not a feature for smaller sellers; it’s a threat.
These roadblocks aren’t temporary. Addressing them requires change at multiple levels: clearer legal frameworks defining liability, consumer education that shifts expectations around AI autonomy, and business models that don’t put a platform’s interests directly at odds with a merchant’s. Until those pieces move, agentic commerce remains a lab demo rather than a market reality.
Why This Matters:
This moment reveals how markets actually work. Raw technological capability isn’t enough. Adoption depends on trust, clear legal accountability, and competitive fairness. If regulation remains unclear and dominant platforms can’t convince independent merchants that agentic systems serve their interests rather than threaten them, innovation stalls—even when the technology works. For investors, entrepreneurs, and policymakers, the lesson is stark: the next wave of AI commerce won’t arrive simply because the software exists. It’ll arrive when the legal, competitive, and trust infrastructure supports it. Right now, none of those pieces are in place.