Purpose and scope: This study explores public moral intuition about artificial intelligence (AI)-enabled beauty apps, which are designed to analyse, evaluate, and modify images of people. A subset of these apps now enables users to virtually “try on” AI-simulated cosmetic surgery procedures. In this study, we focus on attitudes about algorithmic invasiveness—defined as the tendency of AI algorithms to increasingly influence and integrate into our daily lives.
Method: We designed a conjoint survey experiment to assess social acceptability of beauty app functionalities with varying degrees of invasiveness (absent vs mild vs high variants). The AI-enabled functionalities (attributes) were “use of personal data”, “beauty scoring”, “image enhancement”, “product recommendation” and “virtual surgery”.
Results: We analysed 464 survey responses, which showed that functionalities with high degree of algorithmic invasiveness are viewed as significantly less socially acceptable than less invasive functionalities. The two attributes that were deemed problematic irrespective of the degree of invasiveness were “beauty scoring” and “virtual surgery”. For the other three attributes, participants deemed variants with high degree of invasiveness were more problematic than the mild and absent variants. User demographics and attitudes further shape perceptions. Surprisingly, older adults display greater acceptance of highly invasive functionalities compared to younger users. Participants with higher self-objectification or those with positive views of AI show increased tolerance of invasive functionalities.
Conclusion: This empirical study supports the notion that we can elicit moral intuitions about new technologies. Specifically, we gained insight about how participants make normative distinction between degrees of invasiveness.