AI · 6 August 2026
AI Shopping Agents & Amazon: US Appeals Court Ruling Explained
A US federal appeals court has ruled Perplexity AI's shopping agent may operate on Amazon, marking the first federal appellate decision on AI agents acting on behalf of users online.
What happened
A US federal appeals court has overturned an injunction that Amazon had secured against Perplexity AI, ruling that the AI startup's shopping agent may continue operating on Amazon's platform. The court's reasoning centred on a critical distinction: it is the user who accesses Amazon, not Perplexity itself — meaning the startup cannot be held directly liable for actions taken on its behalf by a human customer.
The ruling is reported to be the first federal appeals court decision to address whether AI agents can lawfully act on online platforms on behalf of users. Amazon had sought the injunction to block Perplexity's agent from interacting with its marketplace, but the appeals court found that framing legally insufficient under the circumstances presented.
Why it matters
For anyone designing or operating digital commerce experiences, this ruling introduces a new structural reality: AI agents are increasingly the entity that browses, compares and initiates purchases — not the human customer sitting behind the screen. That shifts the locus of the customer experience from a person navigating a UI to a model executing instructions. Platform operators who have built their conversion funnels, personalisation logic and terms of service around human behaviour will need to reconsider what "the customer journey" actually means when the journey is automated.
From a behavioural economics standpoint, the decision also has downstream implications for choice architecture. If AI agents become normalised intermediaries in shopping, the nudges, defaults and friction points that platforms deploy to influence human decision-making become far less effective — or are bypassed entirely. The competitive battleground shifts from winning the human's attention to winning the agent's selection logic.
The Renascence take
Most commentary on this ruling will focus on the legal precedent or the competitive dynamics between Amazon and AI startups. The more consequential question for CX practitioners is quieter: when an agent shops on a customer's behalf, who owns the relationship — and who is responsible for the experience that results?
The instinct will be to treat AI agents as a distribution channel to optimise for, much as brands once scrambled to optimise for Google's algorithm. That framing misses the point. An agent acting on a user's behalf is an extension of that user's intent — which means the experience the agent encounters (pricing transparency, friction, trust signals) reflects directly back on how the user feels about the outcome. Operators who design for agent-mediated journeys with the same empathy and rigour they apply to human journeys will be better positioned than those who treat agents as bots to be blocked or gamed. The ruling does not resolve the ethical questions; it simply makes the operational ones more urgent.
Sources
This briefing was written by our Newsdesk, synthesising reporting from the outlets below. Follow the links for the original coverage.
More in AI
Stay ahead of CX
Get the signal, not the noise.
The stories shaping customer experience — plus the Journal and Experience Loom — in your inbox.


