74% of Consumers Now Use AI for Shopping: What the Latest Data Means for Your Catalog

TL;DR: Three-quarters of consumers already use AI for shopping, yet most retailers remain invisible to these AI-powered discovery channels. The 805% surge in AI traffic during Black Friday 2025 signals an urgent window: retailers who optimize for AI agents now will capture the $385 billion agentic commerce market by 2030. Those who wait risk becoming background utilities in an agent-controlled marketplace.

Here is a startling paradox: 74% of consumers now use AI tools in their shopping journey at least occasionally, according to a Talkdesk survey of 2025 holiday shoppers. Yet only about one in five consumers have even heard the term "agentic AI" and understand what it means. Shoppers are embracing AI without knowing they are at the forefront of retail's most significant transformation since e-commerce itself.

The Invisible Infrastructure Problem

The disconnect between consumer adoption and retailer readiness creates a massive opportunity gap. Consider the numbers: ChatGPT now processes 2.5 billion daily prompts, with approximately 50 million of those involving shopping queries. That is 50 million daily opportunities for your products to be discovered, compared, and recommended.

Yet most retailers have no strategy for this channel.

📊 Key Stat: AI-driven traffic surged 805% year-over-year on Black Friday 2025, and shoppers arriving via AI were 38% more likely to complete purchases.

Walmart understood this early. By partnering with OpenAI and enabling ChatGPT Instant Checkout, they now capture 20% of referral traffic from ChatGPT, up 15% from July 2025. Their customer service resolution times dropped 40% using AI assistance. Meanwhile, Amazon, by blocking AI crawlers, watched its ChatGPT referral share collapse 18% to under 3%, effectively removing 600 million product listings from AI-powered discovery.

One retailer embraced the future. One is becoming invisible to it.

The Trust Paradox: Using AI While Questioning AI

Here is what makes this moment so critical: consumers are using AI for shopping, but they do not fully trust it yet. According to the IAB's January 2026 study, only 46% of shoppers fully trust AI recommendations, and 89% still verify AI suggestions with other sources before purchasing.

This trust gap is not a barrier. It is an opportunity.

The IAB research reveals that among consumers who use AI for shopping, AI is now the second most influential shopping source, surpassing retailer websites, apps, and even recommendations from friends and family. Only search engines rank higher. And 80% of consumers expect to use AI more for shopping in the coming year.

The retailers who build trust with AI-assisted shoppers now will own the relationship when autonomous purchasing becomes mainstream. The window is 18 months, maybe less.

Is your product catalog optimized for AI discovery? See how Paz.ai helps leading retailers structure their data to win AI recommendations.

The Scale of the Shift

The numbers tell a story of exponential change:

• ChatGPT has 800 million weekly active users, processing 2.5 billion daily prompts
• 50 million daily shopping queries flow through ChatGPT alone
• AI traffic grew 805% YoY on Black Friday 2025
• $67 billion in Cyber Week sales were influenced by AI

The Talkdesk survey found that 89% of shoppers used AI in some form during the 2025 holiday season, with 74% reporting they saved hours because the experience was more efficient. Among those who saved time, 43% estimated AI saved them 1-3 hours, 26% saved 4-6 hours, and 7% saved more than six hours.

This is not a niche behavior. It is mainstream adoption happening faster than mobile commerce did a decade ago.

The Winner-Loser Divide

Salesforce data from Cyber Week 2025 reveals the starkest divide yet:

📊 Key Stat: Retailers with AI agents saw 7x higher sales growth (13% vs 2%) compared to those without.

Shopify reported 7x growth in AI traffic and 11x growth in AI-driven orders since January 2025. Their merchants using ChatGPT Instant Checkout are capturing demand that competitors never see.

Target quietly added a ChatGPT integration with multi-item cart support, including fresh food and drive-up fulfillment options. They now capture nearly 15% of ChatGPT referral traffic.

Etsy leverages its unique product catalog to capture over 20% of ChatGPT referral traffic, with AI-referred visitors spending 32% more time on site and showing 27% lower bounce rates.

The pattern is clear: retailers who make their catalogs accessible to AI agents are capturing disproportionate value. Those who block or ignore AI crawlers are systematically excluded from the fastest-growing discovery channel in retail history.

What Drives Adoption (And What Blocks It)

The IAB research identifies specific triggers that drive AI shopping adoption:

What consumers want from AI shopping:

• 85% say AI recommendations feel personally tailored
• 77% say AI makes them feel more confident in decisions
• Average shopping steps increase from 1.6 to 3.8 after AI interaction (deeper engagement)

What would increase trust:

• 88% want transparency about information sources
• 87% want verified customer reviews integrated
• 75% want explanations of how AI generates answers

Top consumer concerns:

• Data privacy (49%)
• Unwanted subscriptions (44%)
• Overspending (41%)
• Incorrect product selection (38%)

The Protocol Fragmentation Challenge

Here is the complexity retailers face: there is no single standard for AI commerce. Multiple protocols compete for dominance:

Agentic Commerce Protocol (ACP): Developed by Stripe and OpenAI, powers ChatGPT Instant Checkout. Adopted by PayPal, Worldpay, and commercetools.

Model Context Protocol (MCP): Created by Anthropic, now donated to the Linux Foundation's Agentic AI Foundation. Used by Microsoft Dynamics 365, Google, and over 90% of organizations by end of 2025.

Universal Commerce Protocol (UCP): Launched January 2026 by Google, Shopify, Etsy, Wayfair, Target, and Walmart. Endorsed by Stripe, Visa, Mastercard, Adyen, and 20+ partners.

As we discussed in our UCP vs ACP analysis, each protocol requires different integrations, different data formats, and different authentication approaches. A retailer wanting full AI visibility would need to maintain multiple technical implementations, each with ongoing maintenance requirements.

This fragmentation creates both a challenge and an opportunity. Retailers who solve the integration complexity now will have a durable competitive advantage.

The Path Forward

Morgan Stanley projects the agentic commerce market will reach $190 billion in the base case and $385 billion in the bull case by 2030, representing 10-20% of all U.S. e-commerce. McKinsey projects up to $1 trillion in orchestrated revenue from AI-assisted shopping in U.S. B2C retail alone.

The question is not whether this transformation will happen. It is who will capture the value.

Before AI Agents:

• Customers search, browse, compare, buy (4+ sessions)
• Retailers optimize for keywords and SEO
• Success measured by click-through rates

After AI Agents:

• AI discovers, compares, purchases (1 session)
• Retailers optimize for agent protocols and data access
• Success measured by AI recommendation rates

"AI agents are no longer just a cost-saving measure, but an incredible purchase and productivity accelerator in commerce." — Caila Schwartz, Director of Consumer Insights, Salesforce

Frequently Asked Questions

What percentage of consumers currently use AI for shopping?

According to a Talkdesk post-2025 holiday survey, 89% of consumers used AI in some form during holiday shopping, with 74% reporting they saved hours as a result. The IAB reports 38% of U.S. consumers actively use AI for shopping today, with another 52% planning to in the future.

How much has AI shopping traffic grown?

AI-driven traffic grew 805% year-over-year on Black Friday 2025 and 670% on Cyber Monday, according to Adobe Analytics. HUMAN Security measured 1,300% growth in agentic traffic from January to August 2025.

Do consumers trust AI shopping recommendations?

Trust remains mixed. The IAB found only 46% of shoppers fully trust AI recommendations, and 89% verify information before purchasing. However, among AI users, AI ranks as the second most influential shopping source, behind only search engines.

What is agentic commerce?

Agentic commerce is AI agents autonomously browsing, comparing, and purchasing products on behalf of consumers. Unlike chatbots that answer questions, these agents execute complete transactions from discovery to checkout without human intervention.

How are leading retailers preparing for AI shopping?

Walmart partnered with OpenAI for ChatGPT Instant Checkout and captures 20% of ChatGPT referral traffic. Shopify made all stores "agent-ready by default" with their Winter 2025 Edition. Target launched a ChatGPT app with multi-item cart support.

What protocols do AI shopping agents use?

Multiple protocols exist: ACP (Agentic Commerce Protocol by Stripe/OpenAI), MCP (Model Context Protocol by Anthropic), and UCP (Universal Commerce Protocol by Google and major retailers). Each requires different technical integrations. Learn more in our UCP vs ACP analysis.

The Time to Act Is Now

While you are reading this, 50 million shopping queries are flowing through ChatGPT today. Walmart is capturing 20% of that referral traffic. Where is your share?

The retailers who optimize for AI discovery now will own the customer relationship when autonomous purchasing becomes the norm. Those who wait will find themselves increasingly invisible to the largest generation of shoppers, who already expect AI to guide their decisions.

The window is measured in months, not years. SAP Emarsys was not exaggerating: 2025 was likely the last year consumers shopped as they always had. The question is whether you will be visible in how they shop now.

Ready to make your catalog visible to AI shopping agents? Paz.ai provides the infrastructure to connect your products to every major AI commerce platform through a single integration. The retailers acting now are capturing the traffic that invisible competitors will never see.