A new survey of frontline contact center agents arrives with a finding that should give collection operations leaders pause: although every agent surveyed uses artificial intelligence daily, not one considers it essential to their work.
The research, from contact center vendor UJET, polled 250 U.S. agents at mid-market and enterprise firms across financial services, healthcare, retail and other sectors that lean heavily on customer contact. All reported using AI every day. Yet 93% said they could do their jobs without major issues if those tools went dark, 78% described the technology as not transformative, and zero respondents called AI critical to daily success.
For the accounts receivable management industry, where agents operate under the FDCPA, Regulation F and a thickening layer of state requirements, the more telling number may be this: 93% of agents said they feel compelled to verify what AI tools tell them before relaying it to a consumer, and 15% called real-time AI recommendations unreliable or inaccurate. In a collections context, an unverified balance, payment term or required disclosure is not a productivity hiccup. It is potential regulatory exposure.
UJET attributes the distrust to “bolt-on” deployments, where AI is layered onto legacy, often on-premises systems and siloed data sources. The result is tool sprawl: 81% of agents reported juggling more than four applications during a single interaction, and 19% used more than seven. Roughly two-thirds of customers, agents said, complain about repeating information they already gave an automated system before reaching a human.
The case for AI is not absent. Every agent surveyed said it saves time, mostly redirected toward handling more contacts, and 69% credited it with cutting after-call work such as note-taking. But the cost calculus is shifting. Gartner projects that by 2030 the cost per resolution for generative AI in customer service will exceed $3, higher than many offshore human agents, which undercuts the headcount-reduction logic driving much current AI spending.
The throughline for collection leaders is one of supervision, not adoption. Salesforce research finds 88% of customers now rate the experience a company provides as highly as its products or services. Agents are wary enough of AI to check its work. Operators weighing where to point automation, and how much human oversight to keep in the loop, have a data set worth reading.
