Customer Service AI Interactions Make People Talk More Robotically

What happened

New research reported by Digital Information World suggests that people who interact with customer service AI systems, such as chatbots and voice assistants, tend to adopt more mechanical, stripped-down communication patterns themselves — speaking and typing in shorter, more transactional terms, similar to how they might address a machine rather than a person.

The findings point to a subtle but consistent behavioural shift: rather than AI simply becoming more human-like in how it responds to us, the reverse dynamic also appears to be occurring, with customers unconsciously mirroring the flatter, more instructional tone often used by automated service agents.

Why it matters

For experience and service-design leaders, this is a meaningful signal about the two-way nature of human-AI interaction. Most investment in conversational AI has focused on making bots sound warmer, more empathetic and more "human" — but this research suggests the influence runs both directions, and that the design of AI touchpoints may be quietly reshaping how customers communicate more broadly, including in contexts beyond the immediate service interaction.

This has implications for brand tone, employee-customer dynamics, and even data quality: if customers default to terser, more robotic phrasing when talking to AI, sentiment analysis, intent detection and feedback mechanisms built on natural language may need recalibrating to account for this shift, rather than assuming customer language remains a stable, human baseline.

The Renascence take

The headline finding — that people mimic the machines they talk to — is not really about AI at all. It is about a well-established behavioural principle: humans instinctively match the register, pace and formality of whoever (or whatever) they are communicating with, a phenomenon service designers have long observed in human-to-human contact centres, and one now clearly extending into human-to-machine exchanges.

Most organisations are optimising AI to sound more human, when the more urgent design question may be the opposite: what tone does our AI want customers to mirror back? A curt, purely transactional bot doesn't just resolve tickets faster — it may be training your customers to be curt back, flattening the emotional data you rely on to spot frustration, churn risk or delight. The fix isn't more anthropomorphism; it's deliberately calibrating AI's conversational register to the behaviour you want to encourage in the humans on the other end, and then testing for it, not assuming it away.