December 01, 2025
AI Arrived in Retail. Did Better CX Come With It?
AI in retail arrived at full speed. Shoppers now use AI to research products, compare prices, browse reviews and hunt deals, yet a simple question still hangs in the air: has any of this actually made shopping better?
New UK research from CI&T’s Retail Tech Reality Check report found that 61% of consumers have used AI while shopping, with more than half saying they do so regularly. Yet despite this widespread use, 68% of shoppers couldn’t name a single AI-powered retail experience that impressed them.
AI may be embedded into everyday shopping journeys, but from the consumer’s perspective, it hasn’t meaningfully improved the experience.
AI Is Everywhere, Just Not Where Customers Feel It
Retailers are investing heavily in AI, especially on backend functions like forecasting, inventory optimisation, pricing and logistics. Recent industry reviews highlight just how broad and fast-growing retail AI adoption is: the global AI-in-retail market is estimated to grow dramatically in the coming years.
At the same time, many of the publicly discussed AI applications remain technical or operational, like warehouse automation, AI-driven logistics, demand forecasting, dynamic pricing, and backend recommendation engines.
What this means in practice is that AI often delivers operational uplift rather than visible improvements for customers. Supply-chain efficiency, stock optimisation, automated checkout or smart inventory are valuable. However, the day-to-day experience for shoppers (browsing, checkout, stock availability, convenience) often remains unchanged. As a result, AI becomes invisible.
Consumers Want Convenience
CI&T’s survey reveals where consumers stand when it comes to priorities. While 64% of shoppers said they want retailers to use AI to improve shopping experiences, the benefits they list are grounded and practical, such as saving time, easing product discovery and helping secure better prices.
These are not demands for futuristic flash but requests for lower friction. This aligns with broader commentary across retail-tech media. At Tech for Retail Paris last week, we noted that AI’s most valuable contributions will come through personalisation and improved product discovery rather than “cool” features or gimmicks.
In real terms, shoppers just want something that works, reliably, transparently, and in service of convenience.
Trust Remains the Biggest Blocker
If usefulness is one barrier, consumer trust and privacy concerns pose another. Even among shoppers open to AI, many remain wary of how their data is used and whether AI-driven decisions are fair or biased. This makes them hesitant to engage deeply with AI-powered features, which often rely on personal data and behavioural tracking.
Because of this hesitation, adoption at scale, especially for personalisation, recommendations or any experience that touches data, becomes risky. That means even if the backend is “AI-ready,” the front door stays closed for many shoppers.
That scepticism also surfaces in everyday consumer conversations. In a Reddit thread discussing AI in shopping experiences, some users argued bluntly that “consumers are not savvy enough” to confidently navigate an AI-assisted journey.
Back to Basics
Brock Simon, Retail Expert Partner at Bain & Company, said that increased use of AI in retail will help companies become more customer-centric. He said:
“Agentic AI lowers the friction of using data. It brings back the art of retail that endless dashboards took away. AI lets us deeply use customer data while focusing on customers again.”
What retail needs now is better deployment with customer experience at the centre. AI should be used for things like:
- Reliable inventory and real-time availability
- Smarter, faster discovery and search
- Seamless omnichannel journeys
- Transparent pricing and personalisation that’s trustworthy
After all, consumers already use AI. What they’re still waiting for is proof that it makes shopping better.
Until retailers turn technology into visible CX gains, AI will remain background noise, and customer frustration may continue.
