As artificial intelligence voice assistants move further away from flat, mechanical speech patterns and adopt natural, conversational cadences, users are increasingly interacting with them much like they would another person. According to new research from the University of Kansas School of Business, these humanlike design choices do more than just make conversations pleasant—they can actually cause users to overlook potential security risks.

The study, published in the journal Information & Management, explores how giving non-human entities human traits alters consumer behavior and risk perception.

The Psychology of Anthropomorphism in Technology

The research article, titled “Sounds Good and Feels Better — How Anthropomorphism Influences Artificial Intelligence Voice Assistant Use,” investigates the impact of assigning human characteristics—such as attractive appearances, friendly personalities, or natural voices—to artificial intelligence systems like Apple Siri or Amazon Alexa.

Lead researcher Andrew Chen, the Ronald G. Harper Professor at KU, points out that people naturally gravitate toward what feels comfortable. While consumers primarily use artificial intelligence voice assistants (AIVAs) for the sheer convenience of getting quick answers without opening a computer, the underlying technology remains difficult to independently evaluate.

"If an AI system has a voice that is more friendly or has an appearance that is more attractive, you also tend to trust it and forget about the security risk," Chen said. "If you talk to people or something humanlike, you probably feel more comfortable."

How Emotional Appeal Overshadows Cognitive Risk

To understand this phenomenon, Chen and his co-authors—Yixiu Yu and Sisheng Lianga of Central Michigan University, Haoqiang Jiang of California State University Channel Islands, and Rui Sundrup of the University of Louisville—analyzed survey data gathered from 277 AIVA users using structural equation modeling.

The research team evaluated how AI designs influence two distinct paths of human judgment:

  • Affective appraisal: Emotional responses and feelings of comfort or attachment toward the technology.
  • Cognitive appraisal: Rational evaluations, including the assessment of security risks and the consumer's resolve to stop using the product.
  • The interaction effect: Rather than operating in parallel, the study found that affective appraisal interacts quadratically with cognitive appraisal. When feelings of comfort increase, risk perceptions drop at an accelerated rate.

"Many times we use emotions or affected information as the heuristic to judge what’s 'behind the box,'" Chen explained. "If you feel comfortable or you feel attached to the AI system, you reduce the cognitive appraisal part and kind of forget about it."

Implications for Future AI Design

While some consumers remain wary of always-on devices—noting privacy concerns such as accidental recordings—the push toward more emotionally resonant technology continues. Chen believes designers should consider the hedonic aspects of AI alongside standard functional capabilities, though he offers a word of caution about advancing emotional emulation too far.

Evaluating AI across intelligence quotient (IQ), emotional quotient (EQ), and spiritual quotient (SQ), Chen notes that while machines excel at IQ, they still have ground to cover in EQ and SQ—an evolution he suggests should be approached with care.