AI Customer Service: Why Customers Say It's Getting Worse

What happened

New reporting from CX Dive finds that customers increasingly believe artificial intelligence has made their service experiences worse rather than better. The finding runs counter to the pitch many brands have made for AI-driven customer service — that chatbots, virtual agents and automated routing would make support faster and easier — and instead points to a gap between how companies are deploying AI and what customers are actually experiencing when they try to get help.

The coverage does not point to a single company or product failure; rather, it describes a broader pattern across the customer service sector, where AI tools intended to streamline interactions are, in many cases, leaving customers feeling less satisfied than they were with earlier, more human-led service models.

Why it matters

For experience leaders, this is a signal worth taking seriously rather than dismissing as anti-AI sentiment. When customers report that a channel has gotten worse, it usually means the technology has been deployed to cut cost or handling time without sufficiently preserving the things people actually value in a service interaction — being understood, reaching resolution quickly, and knowing when and how to escalate to a human.

The gap between vendor promise and customer perception also has a trust dimension: once a brand's AI experience disappoints, customers may generalise that disappointment across the whole relationship, not just the single channel, making recovery harder than a traditional service failure.

The Renascence take

This is less a story about AI failing and more a story about deployment discipline failing. The technology itself is not the variable most likely to disappoint customers — the design choices around it are.

Most organisations are measuring AI customer service by containment and cost-to-serve, not by whether the customer actually felt helped — and that metric mismatch is exactly what produces headlines like this one. The behavioural reality is simple: people tolerate automation when it removes friction and resent it when it removes agency, especially the ability to reach a human on their own terms. A customer-obsessed operator should treat "time to resolution with dignity" as the design brief for AI service, audit escalation paths as rigorously as they audit the bot's accuracy, and resist the temptation to let efficiency metrics stand in for experience metrics. Get that discipline wrong, and AI doesn't just underperform — it actively erodes the trust it was meant to protect.