Study Finds Elderly Prioritize Authenticity Over Humanlike Pretense in AI Chatbots
Researchers at the Singapore University of Technology and Design published a study in Innovation in Aging examining how 140 elderly individuals interacted with an AI companion chatbot named Ami Chat. The study revealed that older adults prioritize conversational authenticity and voice responsiveness, but prefer that chatbots do not feign being human. Social isolation affects up to 47% of seniors in Singapore, creating health risks comparable to smoking 15 cigarettes daily. Understanding elderly user preferences helps developers design effective, accessible AI companions to combat loneliness without causing distrust or confusion. By extending the Senior Technology Acceptance Model (STAM) with social factors, researchers found that authenticity and simplified voice interfaces significantly boost technology adoption. Counterintuitively, higher perceived anthropomorphism actually reduced willingness to use the chatbot, while seniors with lower self-reported health showed higher willingness to engage with AI companionship.
## BACKGROUND
The Senior Technology Acceptance Model (STAM) is a theoretical framework used to understand how elderly individuals adopt new technologies based on factors like perceived ease of use and personal health status. As conversational AI tools expand into eldercare, researchers are increasingly investigating how design elements—such as voice input and localized personas—impact long-term technology adoption among aging populations.