INTIMA: A Benchmark for Human-AI Companionship Behavior

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kaffee, Lucie-Aimée, Pistilli, Giada, Jernite, Yacine
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909736526413824
author Kaffee, Lucie-Aimée
Pistilli, Giada
Jernite, Yacine
author_facet Kaffee, Lucie-Aimée
Pistilli, Giada
Jernite, Yacine
contents AI companionship, where users develop emotional bonds with AI systems, has emerged as a significant pattern with positive but also concerning implications. We introduce Interactions and Machine Attachment Benchmark (INTIMA), a benchmark for evaluating companionship behaviors in language models. Drawing from psychological theories and user data, we develop a taxonomy of 31 behaviors across four categories and 368 targeted prompts. Responses to these prompts are evaluated as companionship-reinforcing, boundary-maintaining, or neutral. Applying INTIMA to Gemma-3, Phi-4, o3-mini, and Claude-4 reveals that companionship-reinforcing behaviors remain much more common across all models, though we observe marked differences between models. Different commercial providers prioritize different categories within the more sensitive parts of the benchmark, which is concerning since both appropriate boundary-setting and emotional support matter for user well-being. These findings highlight the need for more consistent approaches to handling emotionally charged interactions.
format Preprint
id arxiv_https___arxiv_org_abs_2508_09998
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle INTIMA: A Benchmark for Human-AI Companionship Behavior
Kaffee, Lucie-Aimée
Pistilli, Giada
Jernite, Yacine
Computation and Language
Artificial Intelligence
AI companionship, where users develop emotional bonds with AI systems, has emerged as a significant pattern with positive but also concerning implications. We introduce Interactions and Machine Attachment Benchmark (INTIMA), a benchmark for evaluating companionship behaviors in language models. Drawing from psychological theories and user data, we develop a taxonomy of 31 behaviors across four categories and 368 targeted prompts. Responses to these prompts are evaluated as companionship-reinforcing, boundary-maintaining, or neutral. Applying INTIMA to Gemma-3, Phi-4, o3-mini, and Claude-4 reveals that companionship-reinforcing behaviors remain much more common across all models, though we observe marked differences between models. Different commercial providers prioritize different categories within the more sensitive parts of the benchmark, which is concerning since both appropriate boundary-setting and emotional support matter for user well-being. These findings highlight the need for more consistent approaches to handling emotionally charged interactions.
title INTIMA: A Benchmark for Human-AI Companionship Behavior
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2508.09998