Marketplace without a storefront: how ChatGPT is creating a new shopping economy in the age of AI

From catalogue to dialogue: a new paradigm of product search

The key distinction of ChatGPT’s approach lies in its fundamental rejection of the keyword paradigm in favour of contextual understanding of user needs. While traditional marketplaces require precise formulations such as “red sneakers, size 42, on sale,” the AI platform comprehends complex descriptive queries like “I need a gift for a friend’s 35th birthday, something heartfelt for the home that reminds us of our friendship, budget up to $100.”

This approach effectively addresses one of the fundamental problems of modern e-commerce — the so-called “paradox of choice.” Numerous studies on consumer behaviour show that when faced with too many options (typically more than 20–30), the likelihood of making a purchase decreases significantly. Users experience cognitive overload and often postpone their decision or abandon the purchase altogether. ChatGPT offers a carefully curated and relevant selection of products, saving users’ time and radically simplifying the decision-making process.

The following table summarises key facts about ChatGPT’s new commercial functionality, confirmed by official sources from OpenAI and its partners.

Technological foundation: the architecture of the invisible marketplace

The core of the new commercial ecosystem is the Agentic Commerce Protocol, developed in partnership with the payment system Stripe. This technological architecture includes three key components that ensure the seamless and secure operation of the system:

• Product Feed — a mechanism for direct delivery of structured product data from sellers, ensuring real-time accuracy and relevance of information;
• Checkout API — an application programming interface for deep integration with marketplace backends, allowing orders to be processed without redirecting users to external websites;
• Delegated Payments — a secure payment system using tokenisation technology (Shared Payment Token), which protects users’ financial data.

Of particular note is the strategic partnership with Walmart — one of the world’s largest retailers. This alliance demonstrates that traditional retail giants see ChatGPT not as an experimental technology but as a serious sales channel requiring integration into their core business strategy.

Business model and economic prospects

OpenAI notes that sellers will pay a “small commission” on completed purchases. Although exact commission rates have not yet been disclosed, this model creates an entirely new and potentially vast source of revenue for the company, extending far beyond the traditional ChatGPT Plus and Enterprise subscription models.

The project’s economic potential becomes clear when analysing its basic metrics: 75 million commercial queries per week create a critical mass of demand. Even under conservative assumptions — a 1–2% conversion rate and an average purchase value of $50–70 — the annual transaction volume could reach billions of dollars within the first year of full operation.

Market implications and strategic challenges. The emergence of ChatGPT as a commercial platform poses serious strategic dilemmas for all participants in the e-commerce market. For traditional marketplaces such as Amazon and eBay, it represents a direct threat to their business model, which is based on keyword search and paid advertising. For retailers, it is both an opportunity to reach new audiences and a risk of losing control over the customer experience.

For small and medium-sized businesses, the new approach opens up unique opportunities. There is no longer a need to invest heavily in complex search marketing campaigns — it is enough to provide high-quality structured product descriptions and connect to the platform. However, this also introduces a new dependency on AI algorithms that determine which products are shown in response to specific queries.

Limitations and prospects for development

Despite its impressive start, the system faces objective limitations that will shape its evolution in the near term:

• Functional — the ability to purchase only one product at a time significantly limits the average basket size and user convenience.
• Geographical — availability only in the United States excludes the vast markets of Europe and Asia.
• Integrational — a limited number of connected platforms compared with traditional marketplaces.

However, development is progressing at an exceptionally rapid pace. Plans have already been announced to introduce a shopping cart, expand geographical coverage, and connect additional platforms. The technological foundation, built on an open protocol, allows new sellers and payment systems to be integrated with relative ease.

A new shopping economy is only beginning. ChatGPT is creating not just another marketplace but a fundamentally new shopping economy, where value is generated through precise understanding of user needs rather than the scale of catalogues or aggressive marketing. This calls into question the future of traditional e-commerce models built around the principle that “more choice is always better.”

For businesses, this means an urgent need to adapt to a new reality, where the quality of structured data and the ability to integrate with AI platforms become critical success factors. The technological transformation of commerce has begun — and it challenges not only traditional approaches to online shopping but also the very principles governing the relationship between buyers and sellers. Within the next one to two years, we are likely to witness a redistribution of market share, with success favouring those who can adapt to the economy of conversational shopping.