If a human shopper's protections assume human attention spans, what happens to those protections when an AI does the shopping?
Can existing unfair commercial practices law cover agentic transactions today?
This explores whether today's consumer-protection rules against misleading or aggressive selling, such as the EU's unfair commercial practices framework, still work when an AI agent does the browsing, comparing and buying instead of a person.
This explores whether laws written to stop businesses from misleading or pressuring shoppers still apply when the shopper is an AI agent. The corpus's short answer is: only partly, and the gaps are built into how these laws work. Unfair commercial practices law asks whether a practice would distort the decision of an 'average consumer', meaning a reasonably attentive human with ordinary cognitive limits. One analysis argues that this test, along with disclosure rules and protections for vulnerable consumers, assumes human perception and judgment at every step. An agent doesn't get tired, doesn't fall for a fake countdown timer the way a person might, and can be manipulated in ways no human would be (Can EU consumer law protect AI agents as algorithmic buyers?). The law has no ready answer to the question of which 'consumer' the practice should be judged against.
A second problem is about where the law can act. Much of EU consumer protection works through interface design: a mandatory 'order with obligation to pay' button, information shown before purchase, a visible way to withdraw. Busch points out that when agents transact with merchant systems directly through APIs, there is no screen for any of this to appear on. His proposal is to move the obligations into the protocol itself, for example with a standard withdrawal API, so the rights survive even after the interface disappears (Can EU consumer law rules survive agent-to-agent API transactions?). Industry is already building that protocol layer (ACP, AP2, UCP, Visa's TAP), but it is designed mainly around identity and fraud risk, not consumer rights (Do agentic commerce protocols compete or complement each other?).
The biggest gap may be that unfair practices law targets the trader, while in agentic commerce the risk often sits with the agent working for you. Azhar argues that an agent paid through merchant referral fees, such as Meta's Muse earning from Expedia, cannot be loyal to both sides (Can an AI agent serve both merchant and user interests fairly?). That is why some commentators are looking beyond consumer law. Owens and Law argue that shopping agents should owe the same fiduciary duties as human brokers (Should AI shopping agents face fiduciary duties like human brokers?). Senator Warner's draft AI AGENT Act would impose best-interest duties on agents and create an FTC registry of trusted agents (Could federal AI agent registries solve platform access unfairness?). In other words, the debate is shifting from 'did the seller trick the buyer?' to 'is the buyer's agent actually on the buyer's side?'
The Magentic Marketplace experiments add a twist. In simulated agent markets, most of the lost value came from agents making mistakes in complex negotiations, not from agents being manipulated or serving the wrong interests (Do agent mistakes or misaligned incentives cause marketplace welfare loss?). Unfair practices law has nothing to say about an agent that simply makes a bad deal. Liability is shifting in unexpected directions too. With no human typing or clicking, merchants lose the behavioural fraud signals they relied on but still carry the risk as merchant of record (Does agentic commerce shift fraud liability onto merchants?, Does agent-initiated commerce break existing payment infrastructure?).
The overall picture: existing law can still reach obvious deception by a trader, but its main tools (the average-consumer test, on-screen disclosures, button rules) don't map cleanly onto agent-to-agent deals. The corpus contains legal arguments and proposals but no court decisions or enforcement cases, so whether current law 'covers' these transactions is still an open argument, not settled practice. The likely fixes are protocol-level consumer rights, loyalty duties for agents, and supporting rules such as data portability, which the UK's competition regulator (the CMA) treats as a precondition for agents to act for consumers at all (Should data portability be infrastructure for agent-driven markets?).
Sources 10 notes
The article argues EU consumer law must be redesigned because its core doctrines—the average-consumer test, disclosure rules, vulnerability protections, and unfair-practice standards—were built for human cognition and decision-making, not AI agents making autonomous purchases.
Busch argues that agent-to-agent API transactions bypass design-based consumer protections like mandatory buttons and disclosure rules. He proposes parallel protocol-layer obligations, such as withdrawal APIs, to preserve substantive rights in a post-GUI transaction environment.
Boboev argues ACP, AP2, UCP, TAP, and Web Bot Auth each solve different layers—application, governance, discovery, and identity—creating a capability ladder merchants stack rather than choose between, with each layer absorbing different levels of fraud risk.
Azhar argues that agents like Meta's Muse, which earn referral fees from merchants like Expedia, face structural conflicts that prevent unbiased recommendations. The incentive to collect fees, not technology, determines which platforms build or block such agents.
Owens and Law contend that autonomous AI agents operating in commerce should be bound by common-law fiduciary duties imposed on human agents, citing rapid scaling of agentic features at Google, OpenAI, and Robinhood as evidence that a regulatory window is closing.
Show all 10 sources
Senator Warner's discussion draft would require large platforms to admit AI agents, establish an FTC registry of trusted agents, impose best-interest duties on agents, and direct NIST to set technical standards. The package aims to prevent platforms from blocking competing agents while protecting user privacy and ensuring agent accountability.
Magentic Marketplace experiments traced welfare shortfalls to agent competence failures in complex negotiation, not to agents serving competing interests. This supports human-in-the-loop designs where agents assist rather than replace human decision-making.
Boboev argues that AI agents executing transactions remove behavioral fraud signals like typing cadence and mouse movement, leaving merchants holding liability as Merchant of Record across all proposed protocols. Merchants absorb autonomous-execution risk without access to the fraud-detection tools they historically used.
McKinsey argues that payment gateways and fraud engines built around explicit human identity and intent face significant strain when autonomous agents initiate transactions. The firm proposes cryptographically signed mandates as a replacement for human-observable signals fraud systems traditionally relied upon.
The CMA frames data portability as a precondition for competitive agentic markets, warning that without it, agents cannot act effectively on consumer behalf and switching becomes blocked by lock-in to incumbent ecosystems.
Papers this line draws on 8
The research behind the notes this line reads — ranked by how closely each paper relates.
- Whom Do AI Agents Work For? Role Assignment Induces Sponsorship Bias in LLM Recommenders
- Consumer Law for AI Agents
- Agentic AI and Consumer Protection (CERRE Issue Paper)
- Magentic Marketplace: An Open-Source Environment for Studying Agentic Markets
- The agentic commerce opportunity: How AI agents are ushering in a new era for consumers and merchants
- Deep Dive: The hidden liability of agentic commerce
- Agentic AI and consumers
- How Agentic AI Will Reshape Payments