10 Questions to Fall in Love with ChatGPT: An Experimental Study on Interpersonal Closeness with Large Language Models (LLMs)
Abstract
Large language models (LLMs), like ChatGPT, are capable of computing affectionately nuanced text that therefore can shape online interactions, including dating. This study explores how individuals experience closeness and romantic interest in dating profiles, depending on whether they believe the profiles are human- or AI-generated. In a matchmaking scenario, 307 participants rated 10 responses to the Interpersonal Closeness Generating Task, unaware that all were LLM-generated. Surprisingly, perceived source (human or AI) had no significant impact on closeness or romantic interest. Instead, perceived quality and human-likeness of responses shaped reactions. The results challenge current theoretical frameworks for human-machine communication and raise critical questions about the importance of authenticity in affective online communication.
Study specs
Participants evaluated 10 AI-generated responses to an interpersonal closeness task in a matchmaking scenario, without knowing the responses were AI-generated.
- Institution
- University of Duisburg-Essen
- Discipline
- Social Science
- Sample Size
- N=307
- Study Type
- Experimental Study
- Year
- 2025
- Human Data Platform
- Prolific
- Source
- View Source Google Scholar
Measured Outcomes
Impact of perceived response source (human vs AI) on interpersonal closeness and romantic interest; influence of perceived quality and human-likeness.
Peer Review & Critical Discussion
Potential Selection Bias in 2023 Cohort
The participant pool shows a concerning overrepresentation of users from high-income demographics. Looking at Table 3, we can see that 78% of respondents had annual incomes above $75k, which significantly limits the generalizability of these findings to broader populations.
Non-naive Participants Issue
I've noticed a methodological concern regarding participant naivety. Given that Prolific users often complete multiple studies, there's a real risk that participants had prior exposure to similar experimental paradigms, which could confound the results.
RLHF Applicability to This Study Design
The implications for RLHF training pipelines are understated. If we accept the authors' conclusions about preference stability, this has direct consequences for how we should structure reward model training. The temporal decay effect described in Section 4.2 is particularly relevant.
Verify your expertise to join discussion
Create an account and verify your credentials to participate in peer discussions.