Amazon’s foray into AI-powered chat advertising is generating plenty of user data but not translating into meaningful sales, according to The Information. It’s an early signal that conversational AI ads may not be the commerce shortcut retailers hoped for.
The concept sounds promising on paper: embed product recommendations into AI chat interfaces where users are already asking questions and exploring options. Amazon, with its massive product catalog and ad infrastructure, seemed like the perfect company to crack this. But the early results tell a more nuanced story.
🔍 Why Data Without Sales Is a Problem
Gathering behavioral data from AI chat interactions is valuable, sure. But advertisers don’t pay for insights alone. They pay for conversions. If AI chat ads consistently drive engagement without purchase intent, brands will eventually redirect spend to channels that actually move product.
This matters beyond Amazon. Every major platform is racing to monetize AI assistants:
- Google is weaving ads into its AI Overviews in Search
- Microsoft has experimented with ads in Copilot
- Meta is building AI chatbots across its apps with commerce potential
- Perplexity launched sponsored follow-up questions
Amazon’s early stumble suggests the fundamental challenge isn’t technical. It’s behavioral. When users engage with AI chat, they’re in discovery mode, not buying mode. The mental shift from “tell me about this” to “add to cart” is bigger than advertisers assumed.
🧠 What This Tells Us About AI Advertising
Traditional display and search ads work because they meet users at moments of intent. You search for running shoes, you see running shoe ads, you click and buy. The funnel is short.
AI chat interactions are different. Users ask broader questions, compare categories, and explore. That’s useful top-of-funnel activity, but it sits far from the checkout button. Amazon’s data likely confirms what ad researchers have suspected: conversational interfaces create a browsing mindset, not a buying one.
The irony is thick. Amazon built its ad empire on high-intent product searches. Moving that into a chat format dilutes the very intent signal that made Amazon ads so effective in the first place.
📌 What AI Practitioners Should Watch
- Ad format innovation is needed. Simply dropping product cards into chat responses won’t work. Native, contextual recommendations that feel like helpful suggestions (not interruptions) will perform better.
- Attribution models matter. If chat ads influence purchases that happen later through traditional browsing, current measurement may undercount their value.
- Patience vs. pivot. Early data from a new format rarely reflects mature performance. Amazon has the resources to iterate. Smaller players copying this approach may not.
What Comes Next
Amazon isn’t going to abandon AI chat ads over early results. The data collection alone justifies continued investment. But the company will need to solve the intent gap, finding ways to nudge chat users from exploration toward action without breaking the conversational experience.
For the broader AI industry, this is a reality check. Monetizing AI assistants through advertising won’t be as straightforward as transplanting existing ad models into new interfaces. The companies that figure out native AI ad formats, ones that convert without feeling like ads, will own the next era of digital advertising.
The full report is available at The Information for those wanting the detailed breakdown.