AIToolVerify

—  by

Wolf Huang

in


📑 Table of Contents

In March 2026, a federal judge blocked Perplexity’s Comet AI agent from shopping on Amazon — and in doing so, fired the first legal shot in what may be the biggest shift in ecommerce since mobile shopping.

Welcome to the era of agentic commerce: where AI agents don’t just recommend products — they research, compare, negotiate, and buy on your behalf. And it’s moving faster than anyone expected.

Shopify’s president says agentic traffic to their stores grew 15x in a single year. Forrester predicts 20% of B2B sellers will face AI-agent-led negotiations by end of 2026. Google, PayPal, and Salesforce are all racing to build the infrastructure. If you sell anything online, this isn’t a trend you can afford to ignore.

In this guide, we’ll break down what agentic commerce actually means, walk through the real-world battles already happening, and give you a concrete playbook for preparing your ecommerce business — whether you’re on Shopify, Amazon, or any other platform.

What Is Agentic Commerce?

Agentic commerce is a new paradigm where AI agents autonomously perform commerce tasks — from product research and price comparison to placing orders and handling returns — with minimal human intervention.

Think of it this way:

  • Traditional ecommerce: You search, browse, compare, and buy.
  • AI-assisted commerce (2023-2025): AI helps you write better product descriptions or suggests items. You still make every decision.
  • Agentic commerce (2026+): You tell an AI agent “Find me the best running shoes under $150 with good arch support” — and it researches across multiple stores, compares reviews, checks your past preferences, and either buys for you or presents a curated shortlist.

Salesforce defines it as the inflection point “where the market shifts from passive AI assistance to AI autonomy and action.” According to Deloitte’s agentic commerce guide, by 2030, 25% of global e-commerce sales will be enabled by AI agents, and 55% of digital consumers will start product research using LLM platforms.

The key difference from chatbots or AI assistants? Agency. These systems can take action — access your payment methods, navigate websites, complete checkout — not just give advice.

How AI Shopping Agents Actually Work

AI shopping agents combine several technologies to function as autonomous buyers:

  1. Natural Language Understanding: You describe what you want in plain language, including preferences, budget constraints, and use cases.
  2. Web Browsing & Data Extraction: The agent navigates ecommerce sites, reads product pages, extracts pricing, reviews, and specifications.
  3. Reasoning & Comparison: Using large language models, the agent weighs options against your stated preferences, past behavior, and available data.
  4. Action Execution: The agent can add items to carts, apply coupons, and complete purchases using stored payment credentials.
  5. Memory & Learning: Over time, the agent learns your preferences — preferred brands, sizing, budget patterns, quality expectations.

Current examples include Perplexity’s Comet (now facing legal battles), Google Shopping’s AI agent, Amazon’s Rufus (which operates within Amazon’s ecosystem), and emerging startups building on AI ecommerce tools.

The technology stack typically involves browser automation (like Playwright or Puppeteer), LLM reasoning (GPT-4, Claude, Gemini), and payment integration APIs. Some agents use the Model Context Protocol (MCP) — an open standard that Shopify co-developed with Google — to communicate with stores in a structured way rather than scraping web pages.

Amazon vs Perplexity: The First Legal Battle of Agentic Commerce

The lawsuit that’s defining the rules of this new era started in November 2025 when Amazon sued Perplexity AI over its Comet shopping agent.

What Happened

Perplexity launched Comet as an AI browser that could shop on your behalf across any ecommerce site — including Amazon. The agent would:

  • Log into your Amazon account
  • Browse products based on your request
  • Compare options across sellers
  • Complete purchases using your stored payment methods

Amazon’s argument was straightforward: Perplexity’s agents were scraping Amazon’s website without permission, violating the Computer Fraud and Abuse Act (CFAA), and allegedly taking steps to “conceal” their AI agents so they could continue accessing Amazon’s platform undetected.

The Ruling

In March 2026, a federal judge granted Amazon a preliminary injunction, blocking Perplexity’s Comet from accessing Amazon accounts. The ruling was stayed for seven days to allow Perplexity to appeal.

Why This Matters for Every Seller

This case sets a precedent for the entire agentic commerce ecosystem:

  • Platform control vs user choice: Can you use an AI agent to shop on any website, or can platforms block them?
  • Scraping vs structured access: The ruling pushes the industry toward official APIs and protocols (like Shopify’s MCP) rather than web scraping.
  • Seller visibility: If AI agents bypass traditional product listings and search results, how do sellers get discovered?

The battle isn’t over. Perplexity called Amazon’s stance “a broader threat to user choice and the future of AI assistants.” Expect more lawsuits, more regulation, and rapid development of open commerce protocols throughout 2026.

Shopify’s Bet: Agentic Storefronts and the 15x Traffic Surge

While Amazon is fighting to keep AI agents out, Shopify is doing the opposite — building the infrastructure to welcome them in.

On March 16, 2026, Shopify president Harley Finkelstein made a bold statement: Shopify is preparing for AI shopping agents to change everything.

The numbers back it up: traffic from agentic applications to Shopify stores multiplied 15x between January 2025 and January 2026.

Shopify’s Agentic Commerce Stack

  • Agentic Storefronts: A new feature that lets merchants surface their products to AI conversations everywhere. Set up once, toggle AI platforms on or off, and Shopify Catalog syndicates your products to AI chats automatically.
  • Shopify Catalog MCP Server: An open standard, co-developed with Google, that lets AI agents search hundreds of millions of Shopify products, manage universal carts across multiple merchants, and deliver seamless checkout.
  • Sidekick: Shopify’s own AI assistant for merchants — handling store management, analytics, and customer service tasks.

Finkelstein’s vision: “We’re probably more excited about this particular new era of commerce than we have ever been because we think it’s just going to create so much opportunity — not just for the large merchants, but for the long tail of merchants.”

For Shopify sellers using AI tools, this is a massive signal. The platform isn’t just adding AI features — it’s rebuilding its entire commerce layer for an agent-first world.

Key Players Shaping Agentic Commerce in 2026

The agentic commerce landscape is evolving rapidly. Here are the major players and what they’re building:

Platform Giants

  • Google: Launched the Universal Commerce Protocol (UCP) in January 2026, positioning it as an open standard for AI-agent-led shopping. Their Shopping AI agent is integrated into Search and Gemini.
  • Amazon: Taking a walled-garden approach with Rufus (their in-house AI shopping assistant) while blocking external agents. Amazon wants AI commerce to happen inside their ecosystem.
  • Shopify: The most agent-friendly platform, building open protocols and infrastructure for any AI agent to interact with their merchant base.
  • PayPal: Developing agentic payment infrastructure — the “plumbing” that lets AI agents securely handle transactions.
  • Salesforce: Positioning their Commerce Cloud as the enterprise layer for agentic commerce, with AI agents handling everything from product discovery to customer service.

AI-Native Startups

  • Perplexity (Comet): The most visible (and legally challenged) AI shopping agent. Cross-platform browsing and buying capability.
  • ReFiBuy: A new platform that optimizes catalog data specifically for agentic commerce. Launching broadly in Q3 2026.
  • Phia AI: A shopping agent startup that recently raised seed funding, focused on personalized agent-led shopping experiences.

The Emerging Standard: MCP

The Model Context Protocol (MCP) is emerging as a key standard. Think of it like RSS for AI shopping: a structured way for AI agents to interact with product catalogs, place orders, and manage carts without needing to scrape websites. Shopify and Google are co-developing this, and it could become the default language of agentic commerce.

How Agentic Commerce Affects Ecommerce Sellers

If you sell products online, agentic commerce changes the game in several fundamental ways:

1. SEO Gets a New Layer: “Agent Optimization”

When an AI agent shops for a customer, it doesn’t browse like a human. It doesn’t see hero banners, lifestyle photos, or emotional branding. It reads structured data: product specs, pricing, reviews, return policies, shipping times.

This means traditional SEO is necessary but not sufficient. You now need to optimize for:

  • Structured product data (schema markup, clean titles, accurate specs)
  • Machine-readable reviews and ratings
  • Clear, factual product descriptions (agents care about specs, not vibes)
  • Competitive pricing transparency (agents compare instantly)

Our SEO Survival Guide for 2026 covers the broader picture of how AI is reshaping search — and agentic commerce accelerates every trend we identified there.

2. Price Transparency Becomes Absolute

AI agents can compare prices across dozens of stores in seconds. Markup-heavy pricing strategies become harder to sustain when a buyer’s agent can instantly find the same product cheaper elsewhere. This doesn’t mean you can’t charge premium prices — but you need clear value differentiation that an AI agent can understand and communicate.

3. Brand Loyalty Gets Disintermediated

When an AI agent makes purchasing decisions, brand recall matters less. The agent doesn’t have emotional connections to your brand. It evaluates on specifications, reviews, pricing, and reliability data. To stay relevant, you need:

  • Consistently high review scores
  • Excellent return/exchange data
  • Clear unique selling propositions in your product data
  • Fast, reliable shipping records

4. Customer Service Faces Agent-to-Agent Interactions

Imagine a buyer’s AI agent contacting your store’s AI agent to negotiate a bulk discount, check real-time inventory, or handle a return. This is already starting to happen in B2B (more on that below). AI-powered customer service tools like those in our best AI tools for ecommerce roundup will need to evolve to handle agent-to-agent communication.

B2B Is Next: Agent-to-Agent Quote Negotiation

Perhaps the most disruptive prediction comes from Forrester’s 2026 predictions: 20% of B2B sellers will be forced to engage in AI-agent-led quote negotiations by end of 2026.

Here’s what that looks like in practice:

  1. A buyer’s AI agent sends an RFQ (request for quote) to multiple suppliers simultaneously.
  2. The agent evaluates responses based on price, delivery time, minimum order quantities, quality certifications, and historical reliability data.
  3. It then counter-negotiates — asking for better terms, volume discounts, or faster delivery.
  4. A seller’s AI agent dynamically responds with counteroffers based on margin requirements, inventory levels, and customer history.

Forrester notes that 61% of B2B purchase influencers already say their organization has or will use a private genAI engine to support purchasing. The move from AI-assisted research to AI-led negotiation is a natural next step.

For sellers working on cross-border ecommerce or wholesale, tools like Alibaba’s Accio are early examples of AI-powered sourcing agents that bridge the gap between suppliers and buyers.

What Ecommerce Sellers Should Do Right Now

You don’t need to panic, but you do need to prepare. Here’s a practical checklist:

Immediate Actions (This Quarter)

  1. Audit your structured data. Ensure every product has complete schema markup — price, availability, reviews, specifications, shipping info. AI agents rely heavily on structured data to make decisions.
  2. Clean up your product descriptions. Add clear, factual specs alongside your marketing copy. An AI agent needs to know “100% merino wool, 180 GSM, machine washable” — not just “luxuriously soft.”
  3. Check your Shopify/platform AI settings. If you’re on Shopify, explore the new Agentic Storefronts features and ensure your catalog is synced.
  4. Monitor your reviews aggressively. In an agent-first world, your review score and sentiment become even more decisive. Use AI tools to monitor and respond to reviews systematically.

Near-Term Strategy (Next 6 Months)

  1. Invest in product data quality. Consider tools like ReFiBuy or dedicated PIM (Product Information Management) systems that optimize your catalog for AI discovery.
  2. Build an API layer. If you run a DTC (direct-to-consumer) store, start thinking about how AI agents could interact with your product catalog programmatically. The MCP standard is worth watching.
  3. Rethink your pricing strategy. With absolute price transparency, compete on value bundles, exclusive products, loyalty programs, and service quality rather than just price.
  4. Prepare your customer service for agents. Your chatbot or support system may soon be handling queries from AI agents, not just humans. Make sure your FAQs, return policies, and product information are machine-readable.

Long-Term Positioning (2026-2027)

  1. Own your customer data. As AI agents disintermediate the shopping experience, first-party customer data and direct relationships become your moat.
  2. Consider an “agent-friendly” certification. As standards emerge, being early to adopt open commerce protocols could give you a competitive advantage in agent discovery.
  3. Watch the legal landscape. The Amazon vs Perplexity case will set precedents. Understanding your rights (and obligations) as a platform participant matters.

🐺 Wolf’s Take

I’ve been selling online for over 20 years, and agentic commerce is the most significant paradigm shift I’ve seen since mobile shopping went mainstream.

Here’s what most people miss: this isn’t about technology — it’s about power dynamics. For the last decade, platforms like Amazon and Google controlled discovery. SEO and PPC were the gatekeepers. Agentic commerce redistributes that power to whoever builds the best data layer and the most trustworthy product experience.

The Amazon vs Perplexity lawsuit tells you everything. Amazon isn’t fighting Perplexity because the technology doesn’t work — they’re fighting because it works too well. If an AI agent can shop across Amazon, Walmart, and your DTC store simultaneously, Amazon loses its lock-in advantage. That’s why Shopify is taking the opposite approach — they want to be the platform that agents choose.

For small and mid-size sellers, this is actually a massive opportunity. In the old world, you needed a huge ad budget to compete for visibility. In the agent world, you need excellent product data, genuine reviews, and competitive pricing. A small brand with 4.8-star reviews and clean product specs could beat a big brand with sloppy data — because the AI agent doesn’t care about your brand recognition.

My advice? Start treating your product data like it’s your most valuable marketing asset. Because in 12 months, it will be. And explore the AI tools available for your platform to get ahead of this curve now.

Frequently Asked Questions

What is an AI shopping agent?

An AI shopping agent is an autonomous software program that can research products, compare prices, read reviews, and complete purchases on behalf of a human user. Unlike chatbots that simply answer questions, shopping agents can take real actions — logging into stores, adding items to carts, applying coupons, and checking out. Examples include Perplexity’s Comet, Google Shopping’s AI agent, and Amazon’s Rufus.

Is agentic commerce legal?

The legal landscape is still being defined. The March 2026 Amazon vs Perplexity ruling suggests that platforms can block AI agents that scrape their sites without permission. However, platforms that offer open APIs or support the MCP standard (like Shopify) are explicitly welcoming AI agents. Expect significant legal development throughout 2026-2027 as courts and regulators catch up with the technology.

How will agentic commerce affect my Amazon or Shopify store?

For Amazon sellers, the immediate impact is limited — Amazon is actively blocking external AI agents and keeping shopping within their ecosystem through Rufus. For Shopify sellers, the impact could be significant and positive. Shopify’s Agentic Storefronts feature means your products can be discovered and purchased through AI conversations across multiple platforms. The key for both: invest in clean product data, strong reviews, and competitive pricing.

What AI tools should ecommerce sellers use to prepare for agentic commerce?

Start with the fundamentals: AI product description generators for clean, spec-rich content; AI review management tools; and analytics tools that track how AI platforms are driving traffic to your store. Check our comprehensive best AI tools for ecommerce guide for specific recommendations.

Will AI shopping agents replace human shoppers?

Not entirely, but they will handle an increasing share of routine and considered purchases. Deloitte projects 25% of global e-commerce sales will be agent-enabled by 2030. Think of categories where comparison shopping is high (electronics, commodities, B2B supplies) — those will shift to agents fastest. Experiential shopping (fashion, luxury, artisanal goods) will remain more human-driven, though even there, agents will handle the research phase.

What is the Model Context Protocol (MCP) for ecommerce?

MCP is an open standard co-developed by Shopify and Google that provides a structured way for AI agents to interact with product catalogs, manage carts, and process checkouts. Think of it like an RSS feed for shopping — instead of AI agents scraping websites (which led to the Amazon lawsuit), MCP gives them a clean, authorized data feed. It’s emerging as the default protocol for agentic commerce infrastructure.

Final Thoughts: The Agent Economy Is Here

Agentic commerce isn’t a 2030 prediction anymore — it’s happening right now, in courtrooms, boardrooms, and product launches across the industry. The Amazon-Perplexity lawsuit is defining the legal boundaries. Shopify is building the open infrastructure. Forrester is tracking the B2B adoption curve. And every major tech company is racing to position itself.

For ecommerce sellers, the message is clear: the way customers find and buy your products is about to change fundamentally. Those who prepare now — with clean data, strong reviews, competitive pricing, and platform-ready infrastructure — will thrive. Those who wait will find themselves invisible to the AI agents making more and more purchasing decisions.

The good news? You don’t need to rebuild your business overnight. Start with your product data. Clean it up. Structure it. Make it machine-readable. That single step puts you ahead of 80% of your competitors.

And if you want to explore the AI tools that can help you get there, start with our guides on the best AI tools for ecommerce and AI tools for Shopify.

The agent economy is here. The only question is whether you’ll be ready when the agents come shopping.

Last updated: March 28, 2026