AI in e-commerce: a trend or the new standard?

1. The era of AI agents — how GenAI is reshaping the shopping journey

When Sundar Pichai took the stage at NRF in New York in January 2026 to unveil the Universal Commerce Protocol (UCP) — an open communication standard for AI agents in commerce — few predicted that just a few months later AI agents would become one of the most important traffic-generating channels in e-commerce. The leap from reactive chatbots to autonomous agents that make purchasing decisions on behalf of the customer is transforming the very foundations of the industry.

Today's AI agent does not wait for a command — it works in the background: it monitors prices, analyses reviews, negotiates discounts and finalises transactions. For Gen Z, a generation raised on voice assistants and messaging, handing control of purchases over to an agent is a natural step. Retailers are responding with their own solutions: Home Depot launched "Magic Apron", Ralph Lauren — "Ask Ralph", and Walmart built a next-generation shopping assistant with Google Gemini.

This shift carries profound strategic weight. Gartner forecasts that by the end of 2026, 40% of enterprise applications will include task-specific AI agents. In practice, this means the shopping journey is no longer a linear path from awareness to purchase — it is becoming a network of interactions between customer agents, retailer agents and external platforms. The role of the traditional online store as the shopping interface has begun to erode.

"AI in e-commerce is no longer an add-on. It has become the backbone of sales. 2025 brought a 33-fold surge in AI-driven traffic during Prime Day. AI agents are no longer chatbots — they have begun to make purchases on customers' behalf."

2. The data speaks for itself — e-commerce AI adoption statistics

The numbers leave no room for doubt: 2026 is a breakthrough year for the mass adoption of AI in commerce. Here are the key figures that should be on every board's radar.

• 33-fold growth in AI-driven traffic during Prime Day 2025 compared with 2024 — Adobe Analytics data shows that AI's share of shopping traffic generation is growing exponentially.
• 60% of e-commerce traffic will be generated by AI in 2026 — according to Gartner forecasts, the line between organic search and AI-generated traffic is blurring rapidly.
• 74% of marketers believe AI-powered personalisation increases customer loyalty — according to the Salesforce State of Marketing 2026 report.
• 62% of retailers plan to deploy AI agents in customer service by the end of 2026 — data from the NRF Retail Technology Survey 2026.
• $47 billion — the estimated value of the AI in e-commerce market in 2026 (Statista), representing an 85% increase versus 2024.

Yet the most telling change is qualitative. The first generation of AI in e-commerce (2018–2023) focused on optimisation: better recommendations, smarter search, marketing automation. The second generation (2024–2026) is autonomous action — agents do not suggest, they execute. That is a fundamental difference that is changing the economics of the sales channel.

3. Google Zero and the new reality of search

The most destabilising trend for e-commerce in 2026 is what the industry calls "Google Zero" — a situation in which the user gets an answer without having to click a search result. Google's AI Overviews, ChatGPT Search and Perplexity have changed consumer behaviour: instead of browsing a list of shops, customers ask an agent for a recommendation and receive one, ready-made answer.

The consequences are dramatic. SimilarWeb research shows that the click-through rate of organic search results has fallen by 35% over the last 12 months in e-commerce categories. Gen Z customers skip Google — they compare prices through AI, not through a search engine. That means traditional SEO investment, for years the bedrock of e-commerce, is becoming a risk.

For Polish retailers the implications are particularly painful. Many have built their business models around organic traffic from Google. Google Zero takes away that traffic without offering an alternative attribution model in return. The question arises: how do you reach a customer who no longer searches in the traditional way?

The answer is not to abandon SEO but to diversify channels: build direct customer relationships (loyalty, subscriptions, first-party data), gain a presence in AI-native channels (ChatGPT Actions, Google Shopping Graph), and invest in content that AI agents will cite. This is not a short-term correction — it is a structural change in the architecture of customer acquisition.

4. Risks: privacy, trust, the skills gap

The adoption of AI in e-commerce carries serious risks that are often played down in the Polish context. Three areas require particular attention from boards.

Privacy and data security

AI agents operate on customer data in ways that go beyond traditional consent models. Every agent interaction — price negotiation, product comparison, access to purchase history — is a data exchange that must comply with GDPR while remaining transparent to the user. Research shows that 34% of shoppers are worried about their privacy in the context of AI in e-commerce, and 28% do not trust AI-generated recommendations.

Consumer trust

The paradox of AI adoption is that the more automation, the greater the demand for authenticity. According to the PwC 2026 report, 45% of shoppers still prefer dealing with a human for complex purchasing decisions. Overly aggressive automation — especially in areas requiring empathy, advice or problem-solving — can destroy trust built up over years.

The skills gap

The least visible but most structural risk. The Polish e-commerce market struggles with a shortage of specialists who combine commercial knowledge with AI capabilities. Deploying AI agents requires not only technology but also the skills to design experiences, manage data and build an omnichannel strategy. Without investment in people, AI tools will remain expensive gadgets.

"But 45% of shoppers still prefer a human. And 34% worry about privacy. That is why winning in e-commerce is not about deploying AI. It is about finding the right balance between automation and trust."

5. Recommendations for Polish players

Eight strategic recommendations for the boards of retail and e-commerce companies in Poland, based on an analysis of trends and data from the report "10 trends in e-commerce 2026" (DlaHandlu.pl).

1. Stop treating AI as an IT project. AI in e-commerce is a business strategy. Deploy it within the commercial organisation, not within the innovation department. Agentic AI requires an understanding of purchasing processes, not just coding.
2. Invest in first-party data this year, in 2026. Google Zero and agent-driven commerce are making your own customer data the most valuable asset. Build a unified customer profile that lets your AI agents act with context, not blindly.
3. Prepare an AI-native content strategy. AI agents cite the sources they consider credible. Your product descriptions, returns policies and availability data must be machine-readable and aligned with the formats preferred by Google Shopping Graph and OpenAI Crawler.
4. Build omnichannel around agents, not channels. The customer's AI agent does not distinguish between brick-and-mortar and online stores — it expects consistency. Click-and-collect, ship-from-store and real-time inventory visibility are prerequisites, not innovations.
5. Introduce KPIs for AI. Do not measure success by the number of models deployed, but by concrete indicators: conversion growth, lower service costs, improved NPS, AI's share of generated revenue. Without hard metrics, AI will remain an expensive experiment.
6. Do not ignore risk — manage it. Build an ethical framework and transparency around AI decisions. The consumer has the right to know when they are talking to an agent and when to a human. That is not a legal requirement — it is a competitive advantage.
7. Develop your team's capabilities. The skills gap is the biggest risk facing Polish e-commerce in 2027. Training, hiring AI-native talent and partnering with technology providers is a cost that pays off — provided it is not treated as a one-off expense.
8. Experiment with agent-driven commerce as early as Q4 2026. Run a pilot with an AI agent in customer service or offer personalisation. Scale on the basis of data, not gut feeling. The biggest mistake Polish players made was waiting for mobile-first — do not repeat that mistake with AI-first.

AI in e-commerce is no longer an add-on. It has become the backbone of sales. For Polish retailers and e-commerce players, the next 12 months are the last window to build an advantage before the market consolidates. Those who make decisions in Q4 2026 will make it in time. The rest will be playing catch-up — and in the era of autonomous agents, the gap grows exponentially.

If you are wondering how AI will change your network's strategy — I help retailers design omnichannel that links technology with the real needs of customers. Get in touch, let's talk.