A new research report from Tidio, AI in E-Commerce in 2026: The New Shopping Funnel, exposes a fundamental disconnect in how AI's impact on consumer behavior is measured. Drawing on data from McKinsey, Contentsquare, Similarweb, Bain, and Tidio's own platform, the report finds that AI shapes purchase decisions at a scale that standard attribution models cannot capture.
The core issue, termed 'dark AI,' arises when consumers use AI assistants for product recommendations, then navigate directly to a brand via a new browser tab or branded search. These sessions register as direct or organic traffic, giving the AI no credit. According to McKinsey, half of consumers now rely on AI as their primary source for product research, yet Contentsquare's analysis shows AI-referred sessions account for only 0.2% of retail web traffic.
Conversion data from the few sessions tagged as AI-referred suggests significant undercounting. Similarweb's analysis of U.S. retail data finds ChatGPT-referred sessions convert at 11.4%, the highest rate of any measured channel, ahead of direct traffic (10.2%), paid search (9.3%), and organic search (5.3%). This premium implies that tagged AI referrals represent only a high-intent fraction of a larger pool of AI-influenced journeys.
The attribution gap is widening. TollBit's analysis shows click-through rates from AI applications dropped nearly threefold in 2025, from 0.8% in Q2 to 0.27% by year-end, as AI platforms consume more content while generating fewer outbound clicks.
'Brands making budget decisions based on last-click attribution are optimizing for a measurement system that cannot see what is actually driving demand,' said Tytus Gołas, Founder and CEO of Tidio. 'The inputs that determine AI visibility—feed completeness, structured data, review coverage—live across multiple teams, and no one owns them because no one can see the return.'
The financial stakes are substantial. McKinsey projects $750 billion in U.S. revenue will flow through AI-powered search by 2028, with unprepared brands risking 20-50% of traditional search traffic. Morgan Stanley estimates AI agents will influence $190-$385 billion in U.S. e-commerce spending by 2030.
The report also documents protocol infrastructure being built to formalize AI's role in transactions, including Google's Universal Commerce Protocol, OpenAI's Agentic Commerce Protocol, and Visa's Trusted Agent Protocol. Consumer readiness is accelerating: Omnisend's research found reluctance to allow AI to complete transactions dropped from 66% to 32% between February and July 2025.
The full report is available for download at https://www.getlyro.ai/reports/ai-in-ecommerce. For media inquiries, visit https://www.tidio.com/newsroom.
