Russian e-commerce continues its accelerated integration of artificial intelligence into user scenarios. Following the announcement by Wildberries & Russ, Ozon has begun testing its own AI assistant designed to help shoppers who have not yet made a choice find the optimal product. At this stage, the tool operates on the basis of anonymized platform data and is available only to a limited group of users.

This step is not just a targeted innovation, but a natural stage in the transformation of the entire e-commerce model. Marketplaces are increasingly integrating conversational services directly into the customer journey, aiming to reduce decision-making time and, consequently, increase conversion. Classic filter-based search is giving way to interactive interaction.

How the AI assistant works

At the pilot stage, the assistant is aimed at users who have difficulty making a choice. After switching to a special AI mode, the user finds themselves in a chat where the system asks clarifying questions — about budget, key product characteristics, and the specific task for which it is needed. Based on the answers received, the algorithm generates a personalized selection, taking into account several critically important parameters:

  • Product price.
  • Delivery times and conditions.
  • Seller rating and reliability.
  • Reviews from other customers.

A key feature is operation on anonymized data. The system analyzes aggregated search queries, purchase history, and a body of reviews, as well as current prices, without tying to a specific user identity. This approach reduces privacy-related risks while allowing the algorithm to learn from real demand patterns.

A paradigm shift in search

The introduction of the conversational format changes the very logic of interaction with the catalog. Instead of tediously sifting through product cards and adjusting filters, the user simply describes their need in words, and the system translates it into a specific selection. In essence, this is a shift from classic search results to a personal consultation model, which for the platform is a way to retain an indecisive shopper within the service and guide them to completing an order.

Limited testing allows Ozon to go through the initial verification stage in several areas: the accuracy and relevance of recommendations, the system's resilience to load during dialogue, and the usability of the chat interface itself for the user.

Competitive market dynamics

Ozon's actions fit entirely within the overall trend. Back in late June 2026, its main competitor, the merged company Wildberries & Russ, announced the launch of a similar service. Their assistant, integrated into the mobile app, also helps select products, including gifts, and compare similar items. The competitor's plans include training the system to suggest products from photos and account for purchase history, which is part of a broader strategy to integrate AI into all ecosystem projects.

Thus, the largest Russian marketplaces are moving almost in sync toward the implementation of conversational assistants, competing for user attention at the most important stage — the selection stage. A pilot launch with limited access is a standard and measured approach that allows testing the quality of recommendations and load before scaling to the entire audience.

My view: The synchronized launch of AI assistants by the two giants is a signal that the race for personalization is moving into the practical realm. The success of these pilots will be determined not so much by technological complexity as by the quality of recommendations and how organically the conversational interface fits into familiar user behavior. In the coming quarters, we will witness whose algorithm proves more accurate and whose chat is more convenient, and this will undoubtedly set the development vector for the entire e-commerce market.