Amazon Prime Packages, Prime Day

March 13, 2026

eplisterra/Depositphotos.com

​​Should Amazon Be Banning AI Shopping Agents From its Platform?

Share: LinkedInRedditXFacebookEmail

Amazon won a temporary injunction against Perplexity to block its Comet browser from accessing password-protected parts of Amazon’s website to shop on behalf of human customers.

The ruling sets a precedent for how retailers can defend against unauthorized AI data collection on e-commerce sites, but further legal battles are expected as more AI startups build autonomous shopping and browsing tools.

Amazon sued Perplexity in November, accusing the AI startup of ​covertly scraping data from private Amazon customer accounts through its Comet browser and ​associated AI agen — and of disguising automated activity as human browsing. The lawsuit said ‌Perplexity’s ⁠system posed security risks for customer data. Amazon wrote in its original complaint that Perplexity’s agents “can act within protected computer systems, including private customer accounts requiring a password.”

Amazon also said Perplexity’s agents created challenges for the company’s advertising business, because when AI systems generate ad traffic, the impressions have to be detected and filtered out before advertisers can be billed. Unlike humans who visit shopping sites, AI bots can those bypass ads and sponsored search results.

“This requires modifications to Amazon’s advertising systems, including developing new detection mechanisms to identify and exclude automated traffic,” Amazon wrote in its complaint. “These system adaptations are necessary to maintain contractual obligations with advertisers who pay only for legitimate human impressions.”

Perplexity argued in its opposition to a preliminary injunction that Amazon isn’t so much interested in cybersecurity as it is in eliminating a competitor to its own agentic AI tools.

“This lawsuit is a bald attempt by Amazon to block its own customers from using defendant Perplexity’s groundbreaking Comet AI Assistant on Amazon.com,” Perplexity said. “Why? Because AI agents don’t have eyeballs to see the pervasive advertising Amazon bombards its users with and cannot be upsold to buy more products. Those are the real reasons Amazon filed this suit and seeks a preliminary injunction — not its claimed altruistic concern about protecting consumer data and the ‘customer experience.’”

Perplexity Calls Out Amazon in Appeal

Perplexity appealed the ruling this week, and told CNBC it “will continue to fight for the right of internet users to choose whatever AI they want.”

Amazon said in a statement that the preliminary injunction was “an important step in maintaining a trusted shopping experience for Amazon customers.”

Amazon has broadly locked down its shopping sites from AI agents, blocking dozens of agents — including OpenAI’s ChatGPT — in a “walled garden” while investing in its own AI e-commerce homegrown tools, such as its Rufus shopping assistant.

At the same time, Amazon has started to pull other retailers’ products into its Rufus shopping results, in behavior closely mimicking AI shopping agents. On March 11, Amazon announced it was expanding its Shop Direct program, which allows customers to discover and buy products from other online stores even if those items are not sold on Amazon’s marketplace.

Shop Direct now supports third-party product feeds through services such as Feedonomics, Salsify, and CEDCommerce. The Wall Street Journal noted that other retailers “have shown more willingness to work with AI companies, with boundaries.” Walmart, for instance, last year partnered with OpenAI to enable shoppers to purchase its products directly within ChatGPT, but later clarified that it would use its own AI chatbot, Sparky, inside ChatGPT or other AI platforms for shopping.

An earlier option that enabled shoppers to buy items directly within ChatGPT without visiting Walmart is also being phased out.

BrainTrust

"Retailers should be very cautious about trying to ban or restrict external AI shopping agents outright. The momentum behind agentic commerce is real."
Avatar of Scott Benedict

Scott Benedict

Founder & CEO, Benedict Enterprises LLC


"You know that feeling of hitting a paywall for an article? Then you shrug and go elsewhere? Right now that is a human interaction. AI bots may go elsewhere when blocked."
Avatar of Brian Numainville

Brian Numainville

Principal, The Feedback Group


"Opening the door to external AI agents runs counter to Amazon’s control over the customer journey. Of course, there is a risk that Amazon loses out on agentic search traffic."
Avatar of Neil Saunders

Neil Saunders

Managing Director, GlobalData


Recent Discussions

Discussion Questions

Should retailers be seeking to ban, restrict or work with external AI shopping agents?

What does the ruling mean for consumers and the potential of agentic commerce?

Poll

6 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Neil Saunders

Amazon’s goal is to remain one of the first ports of call for online shopping. That is why it has invested so heavily in broadening its marketplace, sharpening search, enabling discovery through AI agents, and expanding Shop Direct (which allows brands not sold on Amazon to surface in search results). Opening the door to external AI agents runs counter to this aim as it weakens Amazon’s control over the customer journey. Of course, there is a risk that Amazon loses out on agentic search traffic, but it’s betting that its scale and market power are great enough to minimize this.

Last edited 4 days ago by Neil Saunders
Craig Sundstrom
Craig Sundstrom

I think a certain amount of caution is justified, even if this is likely less about security and more about giving them time to turn this to their own advantage (It is their site, after all)

But as Rod Serling might say. Let’s file this under
A for “Amazon” and
I for “Irony”

Last edited 4 days ago by Craig Sundstrom
Gene Detroyer

“It is their site, after all.”

Mohit Nigam
Mohit Nigam

Amazon’s argument about password-protected areas is a serious one. If an AI agent has your login credentials, the surface area for data breaches increases. This ruling reinforces that “unauthorized access” is a legal red line, even if the user “consents” to the AI doing it
Also, Rather than banning agents, retailers should provide official APIs (like Amazon’s Shop Direct or Walmart’s Sparky integrations). This allows agents to fetch data securely without “scraping” or “disguising” themselves as humans
We are witnessing the birth of ‘Agentic Commerce,’ and it’s messy. The real solution isn’t lawsuits; it’s a new technical standard for how AI agents talk to e-commerce platforms. We need a ‘Robot-Exclusion Protocol’ for the 2020s that allows agents to shop while ensuring the retailer gets compensated for the traffic. Until then, we’re just watching two giants fight over who gets to control the customer’s wallet.

Scott Benedict
Scott Benedict

Retailers should be very cautious about trying to ban or restrict external AI shopping agents outright. The momentum behind agentic commerce is real, and consumers will ultimately gravitate toward tools that help them save time, compare options, and make better purchase decisions. As I noted in several recent LinkedIn posts on this topic, the smarter strategy for retailers is not to fight the technology but to learn how to work with it. That means ensuring product data, pricing, and availability information are structured and accessible so AI agents can accurately represent the retailer’s offerings in discovery and recommendation environments.

From a consumer perspective, the emergence of AI shopping agents has the potential to shift the balance of power toward the shopper. These tools can search across retailers, evaluate product attributes, and even transact on behalf of the consumer, dramatically compressing the discovery and decision-making process. If retailers attempt to block these agents, they risk creating friction for customers who increasingly expect digital assistants to help them navigate an overwhelming number of choices.

The more productive path forward is collaboration and governance. Retailers should focus on defining how AI agents interact with their platforms, ensuring accuracy, transparency, and brand integrity while still enabling innovation. In the long run, agentic commerce is unlikely to disappear simply because some retailers resist it. Instead, the companies that benefit most will be those that prepare their digital shelf, product content, and data infrastructure to operate effectively in a world where both humans and AI agents are participating in the shopping journey.

Brian Numainville

You know that feeling of hitting a paywall for an article? And then when you shrug and go elsewhere? Right now that is a human interaction. When an AI agent is shopping for someone and hits that wall, whether Amazon or someone else, it may just shrug and go elsewhere to find what is being sought. Risky business to block potential sales avenues.

6 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Neil Saunders

Amazon’s goal is to remain one of the first ports of call for online shopping. That is why it has invested so heavily in broadening its marketplace, sharpening search, enabling discovery through AI agents, and expanding Shop Direct (which allows brands not sold on Amazon to surface in search results). Opening the door to external AI agents runs counter to this aim as it weakens Amazon’s control over the customer journey. Of course, there is a risk that Amazon loses out on agentic search traffic, but it’s betting that its scale and market power are great enough to minimize this.

Last edited 4 days ago by Neil Saunders
Craig Sundstrom
Craig Sundstrom

I think a certain amount of caution is justified, even if this is likely less about security and more about giving them time to turn this to their own advantage (It is their site, after all)

But as Rod Serling might say. Let’s file this under
A for “Amazon” and
I for “Irony”

Last edited 4 days ago by Craig Sundstrom
Gene Detroyer

“It is their site, after all.”

Mohit Nigam
Mohit Nigam

Amazon’s argument about password-protected areas is a serious one. If an AI agent has your login credentials, the surface area for data breaches increases. This ruling reinforces that “unauthorized access” is a legal red line, even if the user “consents” to the AI doing it
Also, Rather than banning agents, retailers should provide official APIs (like Amazon’s Shop Direct or Walmart’s Sparky integrations). This allows agents to fetch data securely without “scraping” or “disguising” themselves as humans
We are witnessing the birth of ‘Agentic Commerce,’ and it’s messy. The real solution isn’t lawsuits; it’s a new technical standard for how AI agents talk to e-commerce platforms. We need a ‘Robot-Exclusion Protocol’ for the 2020s that allows agents to shop while ensuring the retailer gets compensated for the traffic. Until then, we’re just watching two giants fight over who gets to control the customer’s wallet.

Scott Benedict
Scott Benedict

Retailers should be very cautious about trying to ban or restrict external AI shopping agents outright. The momentum behind agentic commerce is real, and consumers will ultimately gravitate toward tools that help them save time, compare options, and make better purchase decisions. As I noted in several recent LinkedIn posts on this topic, the smarter strategy for retailers is not to fight the technology but to learn how to work with it. That means ensuring product data, pricing, and availability information are structured and accessible so AI agents can accurately represent the retailer’s offerings in discovery and recommendation environments.

From a consumer perspective, the emergence of AI shopping agents has the potential to shift the balance of power toward the shopper. These tools can search across retailers, evaluate product attributes, and even transact on behalf of the consumer, dramatically compressing the discovery and decision-making process. If retailers attempt to block these agents, they risk creating friction for customers who increasingly expect digital assistants to help them navigate an overwhelming number of choices.

The more productive path forward is collaboration and governance. Retailers should focus on defining how AI agents interact with their platforms, ensuring accuracy, transparency, and brand integrity while still enabling innovation. In the long run, agentic commerce is unlikely to disappear simply because some retailers resist it. Instead, the companies that benefit most will be those that prepare their digital shelf, product content, and data infrastructure to operate effectively in a world where both humans and AI agents are participating in the shopping journey.

Brian Numainville

You know that feeling of hitting a paywall for an article? And then when you shrug and go elsewhere? Right now that is a human interaction. When an AI agent is shopping for someone and hits that wall, whether Amazon or someone else, it may just shrug and go elsewhere to find what is being sought. Risky business to block potential sales avenues.

More Discussions