Customer preference for human support has shifted decisively against AI-driven service across North America between October 2025 and April 2026. A 6,000-person study spanning the US, UK, and Canada reveals that preference for speaking to a real person increased from 83% to 85%, whilst those favouring AI dropped from 7% to 5%. Frustration with AI agents rose from 54% to 59%, and 31% of customers now say they would hang up if connected to a bot. Simultaneously, trust metrics have deteriorated: 57% report their trust in a business would decrease if it relied predominantly on AI for customer service (up from 53%), and 70% believe service quality would decline if humans were removed entirely. The data exposes a fundamental misalignment between enterprise investment trajectories and customer expectations—organisations continue deploying automation at scale whilst their customer base actively rejects it.
The implications for CX teams are substantial and demand strategic recalibration. This backlash directly challenges the automation-first playbooks that have dominated platform roadmaps at Zendesk, Salesforce Service Cloud, and similar vendors over the past two years. For teams already running Agentforce or comparable AI-native solutions, the question becomes whether these implementations should be repositioned as agent-augmentation tools rather than replacement mechanisms—a fundamental shift in deployment philosophy that requires rethinking workflows, training, and success metrics. The research indicates that 73% of customers would be more loyal to companies using real people for all service interactions (up from 69%), suggesting that human-centric positioning has become a competitive differentiator rather than a cost centre argument.
The deeper concern centres on what customers actually value: understanding and empathy. The data shows declining comfort with AI broadly, rising data privacy concerns, and beliefs that AI diminishes creativity and problem-solving—issues that no amount of prompt engineering or model refinement will resolve. For support leaders, this signals that the ROI conversation around AI must shift from efficiency gains to enablement: how does this technology make your agents faster, more informed, and more capable of delivering the personalised service customers increasingly demand? The organisations winning in this environment will be those treating AI as infrastructure for human excellence rather than a replacement for it.
AI Backlash Grows Across US, UK, and Canada: More Customers Reject Bots for Human Support in 2026 StreetInsider
AI Backlash Grows Across US, UK, and Canada: More Customers Reject Bots for Human Support in 2026 The AI Journal