AMES, Iowa — Imagine losing your credit card, logging into your bank's website for help and being greeted by a virtual customer service representative who looks and sounds exactly like actor Hugh Jackman.
The person helping you isn't actually the Hollywood star — it’s his AI-powered digital twin. And new research suggests that familiar face could significantly improve how you perceive your customer service experience.
“When people see the familiar face of a famous person they like, there’s a halo effect,” said Lingyao (Ivy) Yuan, associate professor of information systems and business analytics in Iowa State University’s Ivy College of Business. “The positive feelings they have about that person can carry over to how they view the service, even before it begins.”
In a study published in the Journal of Management Information Systems, Yuan and her co-authors examined a growing technology trend known as “Celebrity-as-a-Service,” an approach in which companies use AI-powered or human-operated digital replicas of celebrities to interact directly with customers.
“We found customers may be more trusting of a service, more willing to use it and even more forgiving of mistakes when a virtual representative resembles a celebrity they already know and like,” Yuan said.
Yuan said the study’s findings also offer a glimpse into a future where AI-powered “digital humans” become a routine part of everyday business.
“We're moving toward a world where interactions with digital humans will be increasingly common,” Yuan said. “The question isn't just whether the technology works. It's how people respond to it and how those responses shape the decisions they make.”
The next celebrity endorsement is customer service #
Customers have long seen celebrities endorsing products in commercials and advertisements, but digital twins take that idea much further, Yuan said.
“Instead of appearing in a one-way marketing campaign, a celebrity's digital likeness can interact with customers in real time, answer their questions, solve problems and complete transactions,” Yuan said.
To examine how everyday customers react to digital twins, Yuan collaborated with fellow researchers Antino Kim, associate professor of information systems at Indiana University; Mike Seymour, a researcher in digital innovation at the University of Sydney; and Alan R. Dennis, distinguished professor of information systems and the John T. Chambers Chair of Internet Systems at Indiana University.
The team’s question was simple: If two digital customer service representatives provide the same help, does it matter whether one looks like a celebrity?
To find out, the researchers created a digital prototype, in compliance with U.S. fair use law for research, using technology that placed actor Hugh Jackman's face onto a digital representative. Yuan said Jackman was selected because he combined high name recognition with a largely positive public image, and his diverse career across film, television and theater made him familiar to a wide range of participants.
“This helped us isolate the effects of perceived celebrity status rather than familiarity with a specific type of performer,” Yuan said.
The researchers conducted two controlled experiments in which participants interacted with either the celebrity version or a non-celebrity version of the representative, both of which provided the same assistance. Afterward, participants evaluated the experience across several measures, including whether the representative appeared “capable, benevolent and honest.”
Yuan said the celebrity advantage was clear.
“Participants consistently rated the digital celebrity agent more favorably across each of these measures, even though the quality of service never changed,” she said.
The results highlighted a powerful psychological effect, Yuan said.
Rather than judging the technology solely on its performance, “participants appeared to transfer some of their positive feelings about the celebrity to the service itself,” she said. “We know people often make quick judgments based on social cues, and a familiar, well-liked face can create positive expectations and boost trust.”
A cushion when AI gets things wrong #
The researchers also wanted to know what happens when technology makes mistakes.
In a second experiment, Yuan and her team introduced common customer service errors, such as mishearing a name, misunderstanding an address or asking customers to repeat information.
Typically, those mistakes erode trust and increase frustration, but participants interacting with the celebrity version were noticeably more forgiving and less quick to anger, Yuan said.
"The positive feelings attached to the real-world celebrity acted as a psychological buffer against errors and softened customer judgment," Yuan said.
The researchers also didn’t find any evidence of “algorithm aversion,” Yuan added, meaning participants weren’t more critical when a celebrity avatar was controlled by AI rather than by a human.
Why it matters #
The study’s findings could help companies design more effective digital customer service experiences, but Yuan said they also raise important questions for customers.
As AI-powered "digital humans" become more common, companies may be able to customize virtual representatives for individual customers, “matching celebrities or digital assistants to a user's specific age, language or personal preferences.”
Yuan said transparency and ethical guidelines will become increasingly important as the technology evolves, helping ensure customers always know when they are interacting with AI rather than an actual person.
"Customers need to be aware that there is artificial intelligence behind it," she said. "This technology has the potential to offer genuine value when used in a regulated and ethical way, and as our exposure to it grows, it’s important that we continue to build our shared awareness and understanding of it.”
Yuan said the study also suggests that when it comes to AI, appearance matters.
“A familiar face may not change how well a service works, but it can change how people feel about using it,” she said.
Contacts #
- Lingyao (Ivy) Yuan, Information Systems and Business Analytics, [lyuan@iastate.edu](mailto:lyuan@iastate.edu) , 515-294-7228
- Lisa Schmitz, News Service, [lisaschm@iastate.edu](mailto:lisaschm@iastate.edu) , 515-294-0704
Read the paper #
Yuan, L. (Ivy), Kim, A., Seymour, M., & Dennis, A. R. (2026). Celebrity-as-a-Service: Trust and Willingness to Use Digital Twins for Customer Service. Journal of Management Information Systems, 43(2), 335–367.