AI systems deployed across customer journeys now operate at speeds that outpace human comprehension, creating what Patricia Camden of EY terms the "human tempo gap"—a mismatch between machine velocity and customer ability to process, understand, and trust decisions being made on their behalf. CX leaders have long optimised for friction reduction and speed, a logic that remains sound for transactional tasks like password resets or order lookups. Yet AI fundamentally alters this calculus by enabling systems to infer intent, pre-populate information, generate recommendations, and advance journeys before customers have fully considered their options. The operational metrics may signal seamlessness, but customers experience tempo shock—the disorienting sensation of asking "what the heck just happened?"—particularly in high-stakes moments involving financial commitments, health decisions, sensitive data, or major life changes. An airline rebooking a passenger swiftly without visibility of alternatives exemplifies this gap: the speed reads as efficiency to the system, frustration to the customer.
The implications for CX teams are substantial and demand a recalibration of success metrics. Traditional KPIs—conversion rates, task completion, unsubscribe rates—obscure the quieter signals that reveal genuine friction: customers double-checking information, pausing after recommendations, or abandoning self-service channels to seek human assistance. A click does not equal confidence, and a conversion does not equal trust. This distinction becomes critical as teams implement AI-powered personalization and journey orchestration tools; the data may suggest a customer is primed for outreach or a recommendation, yet immediate action can feel intrusive and reveal the extent of behavioural monitoring, eroding rather than building trust. For teams already running systems like Zendesk's outcome-based pricing models or similar AI-augmented platforms, the challenge is identifying where intelligent friction—deliberate moments of clarity and control—should replace pure speed optimisation.
The strategic question facing CX leaders is whether their AI deployments are genuinely useful or merely presumptive, whether they create confidence or discomfort, and critically, whether they provide customers an easy path to redirect or reject the experience. Camden argues that sometimes the right answer is not the fastest possible message, and teams must test whether their AI-driven actions serve customer agency or merely maximise throughput. This tension becomes more acute given the broader industry trend of cutting human oversight in response to AI failures—precisely when human judgment is needed most to calibrate when speed serves customers and when it alienates them.
Francesca Roche sits down with Patricia Camden, US Marketing, Customer Experience and Loyalty Leader at EY, to examine a growing risk for brands deploying artificial intelligence across customer journeys: AI can act faster than people can process, understand, and trust what is happening. For years,