AI assistants are moving from answering questions to filling shopping carts, as OpenAI and Perplexity race to turn conversational search into a new kind of e‑commerce gateway. Their timing is aggressive, but the rest of the retail and search ecosystem is reacting with more curiosity than panic, signaling that the incumbents believe their own data moats and ad businesses still give them room to breathe.
I see this moment less as a sudden disruption and more as the start of a long contest over who owns the “last mile” between product discovery and purchase, with OpenAI and Perplexity betting that natural language interfaces can pry open a space that traditional search, marketplaces and affiliate publishers have not fully captured.
AI shopping arrives as a feature, not a revolution
The most striking thing about the new AI shopping tools is how quietly they slot into existing user behavior. OpenAI is not launching a standalone marketplace, it is weaving product recommendations and buying guidance into the same chat interface people already use for coding help and travel planning, turning ChatGPT into a kind of ambient shopping companion rather than a destination store, as detailed in coverage of its AI-powered shopping rollout. Perplexity is following a similar path, treating commerce as an extension of its answer engine instead of a separate tab, which keeps the experience feeling like search that happens to end in a purchase.
That framing matters because it lowers the bar for adoption: users do not have to learn a new app or remember a new URL, they simply ask more specific questions about what to buy, from “best 2025 hybrid SUV for city driving” to “which mirrorless camera under $1,500 has the fastest autofocus.” Perplexity’s own product blog leans into this behavior, positioning its tools as a way to “shop like a pro” by turning vague intent into structured comparisons, and then surfacing curated options inside the same conversational thread, a flow it spells out in its shopping assistant guide.
Perplexity’s shopping push targets the search funnel
Perplexity’s strategy is to attack the messy middle of the search funnel, where people bounce between review sites, Reddit threads and retailer pages before deciding what to buy. The company has added dedicated shopping features to its core answer engine so that when a query has clear commercial intent, the interface pivots into product cards, side‑by‑side specs and pros‑and‑cons summaries, a shift described in detail when the startup added shopping features as search competition tightened. Instead of ten blue links, users see synthesized guidance that tries to compress an hour of research into a few scrolls.
To make that useful, Perplexity is leaning on partnerships and payments infrastructure that can close the loop from recommendation to checkout. Reporting on its commerce roadmap notes that the company is working with established payment players so that users can move from an AI‑generated shortlist to a completed transaction with fewer redirects, a direction underscored when coverage of its online shopping plans highlighted how it is integrating with PayPal-powered flows to streamline purchases. The result is less like a traditional search results page and more like a guided buying session that happens to live inside a search box.
OpenAI turns ChatGPT into a commerce concierge
OpenAI’s move into shopping is more explicitly framed as a challenge to the giants of e‑commerce and search, but the mechanics are similar: use conversational context to infer what a shopper actually needs, then surface a small set of options instead of an overwhelming catalog. Coverage of the launch describes how the company is layering product discovery into ChatGPT so that a user planning a ski trip can ask for gear recommendations, get a tailored packing list, and then click through to buy, all within a single flow that positions the assistant as a shopping-oriented concierge rather than a generic chatbot. The commercial intent is clear, but it is wrapped in the same conversational tone that made ChatGPT mainstream.
What differentiates OpenAI’s approach is the scale of its existing user base and the halo effect of the ChatGPT brand, which marketers already see as a new front door for product discovery. Industry commentary points out that “everyone is talking about ChatGPT shopping” in agency circles, with strategists dissecting how to get their brands into the recommendations stream and whether this will cannibalize or complement search ads, a sentiment captured in a widely shared LinkedIn discussion among media planners. That buzz gives OpenAI leverage with retailers and affiliate partners, even as the company experiments with how aggressively to monetize the new behavior.
Free agents and paid assistants: business models diverge
Under the hood, OpenAI and Perplexity are testing different business models for AI‑driven commerce, and those choices will shape how aggressively they push shopping in front of users. Perplexity has positioned its shopping agent as a free feature, accessible without a subscription, in order to grow usage and gather data on what people actually buy when guided by an AI, a stance laid out in coverage of how it launched a free shopping agent to challenge rival assistants. That decision suggests Perplexity is betting on downstream revenue from affiliate fees, sponsored placements or transaction partnerships rather than gating commerce behind a paywall.
OpenAI, by contrast, is threading shopping into a product that already has a tiered pricing structure, which gives it more levers to experiment with premium shopping features for paying users while keeping basic recommendations available to the broader audience. Analysts following the rollout note that the company can treat commerce as one more capability that makes its subscription tiers more attractive, even as it explores direct revenue from retailers who want preferred placement in ChatGPT’s suggestions, a dynamic that has already drawn interest from agencies tracking how AI search engines go on a shopping trip with brand budgets. The contrast is stark: one player is using free access to build a commerce graph, the other is folding shopping into an existing monetization stack.
Consumers get guidance, not just links
For shoppers, the most immediate change is qualitative rather than structural: instead of sifting through dozens of tabs, they can ask for help in plain language and get a synthesized answer that feels closer to a trusted friend’s advice than a search results page. Reviews of early AI shopping tools emphasize how they help people who are “struggling with that buying decision” by turning fuzzy needs into concrete trade‑offs, such as whether to prioritize battery life over camera quality in a mid‑range smartphone, a benefit highlighted in coverage of an AI shopping tool that could help indecisive buyers. That shift from information overload to guided choice is the core value proposition both OpenAI and Perplexity are chasing.
At the same time, the assistants are starting to handle more of the grunt work that used to fall on comparison sites and human sales associates, such as filtering by very specific constraints or explaining technical jargon in context. Industry observers note that Perplexity’s shopping features, for example, can parse a query like “best 2024 OLED TV for a bright room under 1,200 euros” and return a short list with explanations tailored to that scenario, a capability that has been flagged in analyses of how its AI-powered shopping tools reframe product research. The assistants are not just pointing to where information lives, they are interpreting it on the shopper’s behalf.
Retailers and rivals watch, but do not flinch
Despite the fanfare around AI shopping, the reaction from big retailers, marketplaces and search incumbents has been measured rather than alarmed. Reporting on the competitive landscape notes that while Perplexity’s new features tighten the race in AI search, they still sit on top of product feeds and affiliate programs controlled by established players, which limits how much immediate disruption they can cause, a reality underscored when analysts framed its launch as one more step in a tightening search competition rather than a wholesale reset. Amazon, Google and major retailers continue to invest in their own AI layers, confident that their control over inventory, logistics and ad demand gives them a defensive moat.
Marketers, too, are experimenting without abandoning existing channels, treating AI assistants as an incremental surface where product feeds and performance budgets might eventually flow. Media agencies tracking the space describe a test‑and‑learn posture, with some brands feeding structured product data into these assistants while keeping the bulk of their spend in search and marketplace ads, a cautious stance reflected in roundups that place Perplexity’s launch alongside other AI shopping tools and other news rather than as a singular tipping point. The consensus so far is that AI shopping is important to watch, but not yet a reason to rewrite entire media plans.
Media, agencies and the next phase of AI commerce
Where I see the most immediate impact is in how media companies and agencies think about their role in the shopping journey. If AI assistants are going to sit between consumers and retailers, then the value of high‑quality product journalism and expert reviews may shift from driving direct clicks to informing the models that power these recommendations. Trade coverage already hints at this, noting that Perplexity’s shopping tools lean on a mix of structured product data and editorial content to generate their answers, a blend that has been dissected in analyses of how its AI search engine handles commerce queries. The incentive for publishers will be to make their content legible to these systems, not just to human readers.
Agencies, meanwhile, are positioning themselves as translators between brands and AI platforms, advising clients on how to structure product feeds, optimize metadata and negotiate early partnerships with the likes of OpenAI and Perplexity. Industry newsletters that track these moves frame AI shopping as part of a broader shift in how discovery works, grouping Perplexity’s rollout with other experiments in retail media and conversational search, as seen in coverage that highlights its AI shopping tools alongside retailer ad network updates. The next phase will not just be about who builds the smartest assistant, but about who can orchestrate the ecosystem of data, content and commerce that feeds it.
More from MorningOverview