Human First, Machine Second: Perspectives From the Age 85+ on Artificial Intelligence

TL;DR

This study explores how people over 85 engage with and perceive artificial intelligence, highlighting their openness to learning about AI and concerns about privacy and bias.

Contribution

The study provides novel insights into AI perceptions among the age 85+ demographic, emphasizing their unique needs and concerns.

Findings

Participants showed low familiarity with AI but openness to learning about its potential in health and care management.

Participants expressed concerns about AI surveillance, data security, and reinforcing social biases like ageism.

Participants preferred AI that enhances human capabilities rather than replacing them.

Abstract

Rapid advancements in the capabilities of artificial intelligence (AI) have sparked interest in its potential to support an aging population. Minimal work has examined how the oldest of older adults engage with and think about AI in their lives, despite this demographic being one that might especially benefit from AI technologies. Leveraging data from focus groups (N = 14) and a questionnaire (N = 25), this study examines a sample of the age 85+ and their attitudes toward, experiences with, and trust in AI. Participants were recruited from the MIT AgeLab’s 85+ Lifestyle Leaders Panel, a longitudinal study of octogenarians and nonagenarians. Initial findings indicate participants had low familiarity and experience with AI; greatest levels of familiarity were around AI-powered voice assistants. While participants expressed limited experience, many showed openness to learning about AI’s potential to support health and care management, cognition and memory, technology troubleshooting, and personal security. Participants preferred AI that enhanced rather than replaced human capabilities. Their reported concerns about AI centered on its surveillance capabilities, data security, and potential to reinforce social biases, particularly ageism. These findings underscore opportunities to better educate the age 85+ about AI, the advantages of involving them in AI design, and the need to consider how AI may contribute to or bridge digital divides within this population.