UK shoppers distrust AI recommendations, ThoughtSpot finds

ThoughtSpot has published research showing that only one in 10 UK consumers trust AI-powered shopping recommendations. The study also found that most UK shoppers view some forms of personalisation as intrusive.

Based on a YouGov survey of 2,216 UK adults, conducted as part of a wider UK and US sample, the findings point to a gap between retailers' use of AI tools and consumers' willingness to accept them in day-to-day shopping.

In the UK, 76% of respondents said retailers collect too much personal data, while just 4.1% disagreed. The results suggest privacy concerns remain central to how shoppers judge AI-led services, even as retailers place more emphasis on tailored offers, recommendations and automated customer engagement.

Trust in those systems appears weak. Only 10.6% of UK consumers said they trust AI brand or product recommendations when deciding what to buy, while 53% said they actively distrust them. Another 66% said AI had misread their needs or offered an incorrect recommendation.

Privacy concerns

The strongest objections came when personalisation extended beyond a retailer's own site or app. Some 62% of UK consumers said it became too intrusive when their behaviour was tracked across other websites or apps, and the same share objected to real-time location tracking.

ThoughtSpot found that 94% of consumers considered at least one form of retail personalisation intrusive. This points to broad unease over how retailers gather and use behavioural data to influence purchasing decisions.

At the same time, the results show consumers are not rejecting AI outright. About 31% supported AI personalisation when it helped them find better deals, and 19.4% said saving time was a priority when using AI shopping tools. The figures suggest shoppers may accept automation when its purpose is clear and the benefit is immediate.

That balance between usefulness and overreach is emerging as a central issue for retailers. The survey indicates consumers are more open to AI when it solves practical shopping problems than when it tries to predict behaviour in ways they cannot see or control.

Clear limits

Respondents also set out what might make them more willing to trust AI systems. Clear data transparency was cited by 35.2% of consumers, 34.8% wanted simple opt-out controls for AI personalisation, and 31% wanted clear explanations for recommendations.

Those demands reflect a preference for straightforward information and user control over invisible automation. For retailers, the challenge may be not only to improve recommendation systems, but also to explain how they work and what data sits behind them.

The study also linked poor AI performance to commercial risk. Nearly two-thirds, or 63%, said they were unlikely to continue shopping with a retailer if AI regularly suggested products outside their price range. Separately, 59% said they would stop shopping with a brand after a significantly late delivery.

These results place AI recommendations alongside more traditional retail failings such as fulfilment problems. Shoppers appear to judge digital tools by the same standards as pricing, stock accuracy and delivery reliability.

Retail pressure

The findings come as retailers in the UK and other markets increase investment in AI for merchandising, customer service, search, promotions and loyalty programmes. Many of these systems rely on personal data to predict likely purchases or tailor what each shopper sees.

Yet the survey suggests more data collection does not automatically lead to greater trust. If anything, consumers seem prepared to draw a line when tracking goes beyond what they consider necessary for a transaction or a clearly understood service.

There was also evidence of a sizeable middle group of consumers who are not fully opposed to AI but are not ready to embrace it without changes. ThoughtSpot identified 35.2% as a neutral group that said data transparency would help them trust AI-powered recommendations more.

This may be significant for retailers trying to increase adoption of AI-led shopping features without triggering a backlash over surveillance or poor targeting. The commercial opportunity may lie less in increasingly complex personalisation and more in making systems accurate, limited and easy to switch off.

Mayank Sinha, Senior Director, Business Value Consulting, ThoughtSpot, commented: "In my conversations across the industry, retail leaders repeatedly call out that trust is the new battleground for AI adoption. You can't build a reliable customer-facing AI agent on fragmented inventory or opaque tracking.

"Winning the sceptical shopper requires bridging the gap between front-end personalisation promises and back-end operational facts."