How does AI make us feel about ourselves?

By Dr. Tim Sandle
Published September 4, 2023

Computerised technology. Image (C) Tim Sandle

People are happier hearing good news from humans than from AI, but their emotions around bad news are the same regardless of who—or what—delivers it, according to new research from The University of Texas at Austin.

The study was led by Gizem Yalcin Williams, assistant professor of marketing at The University of Texas McCombs School of Business. Williams has been studying people’s reactions towards AI for the past seven years.

The study contradicts conventional wisdom that companies should avoid using AI because they are worried about customers’ reactions when things go wrong. The study probes a seldom-discussed facet of humans’ relationship with artificial intelligence: How do the technologies make us feel about ourselves?

In the study, Williams and her team evaluated more than 5,000 participants across 10 studies. They asked: How do people react to decisions and feedback from AI and automated technologies? And: How can businesses best incorporate these technologies into their services and business processes to maximize customer and employee satisfaction?

They found when people applied for a loan or benefit, those who receive a positive decision from a person felt more joy than did those who receive the very same decision from AI. They were, however, equally unhappy about receiving a negative decision from a person or an AI.

Discussing the results from several recently published studies, researchers further examined situations whereby automated technologies, such as AI, impact various business success factors, including sales, word-of-mouth, employee commitment, and customer satisfaction.

This indicated that when AI or automated technologies are adopted to evaluate and provide feedback for customers or employees, companies should consider adding in some active—not passive—human involvement in those processes and make that involvement clear to customers or employees.

Furthermore, it is apparent that companies that use AI interfaces to interact with customers or employees should consider humanizing those interfaces (e.g., adding an avatar or giving AI a name).

The overview of research and real-world practice offers advice for companies about service and business process design, product design, and marketing and communications.

For example, managers should be selective about when and how they use their more-expensive human workforce for decision-making. For example, since people tend to react the same way to negative news, whether it comes from a person or from AI, companies may consider saving employees’ time for delivering good news.

In designing products that help customers express their beliefs and personalities, companies should anticipate resistance to technological enhancements and consider retaining some element of the product that will make customers feel involved and proud of themselves.

The research has been published in Harvard Business Review, titled “How AI Affects Our Sense of Self.”