When Are LLM Inferences Acceptable? User Reactions and Control Preferences for Inferred Personal Information

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Monteiro, Kyzyl, Park, Minjung, Ioffrida, Alexander, Sanna, Angelina, Hao-Ping, Lee, Mireshghallah, Niloofar, Wang, Yang, Das, Sauvik
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866916000088195072
author Monteiro, Kyzyl
Park, Minjung
Ioffrida, Alexander
Sanna, Angelina
Hao-Ping
Lee
Mireshghallah, Niloofar
Wang, Yang
Das, Sauvik
author_facet Monteiro, Kyzyl
Park, Minjung
Ioffrida, Alexander
Sanna, Angelina
Hao-Ping
Lee
Mireshghallah, Niloofar
Wang, Yang
Das, Sauvik
contents Ask ChatGPT about vacation planning, and it may infer your income. Ask it about medication, and it may infer your medical history. Because such inferences can expose more information than users intend to reveal, prior work argues that they are a defining privacy risk of LLM-based systems. Yet prior work has mostly shown that LLMs can make potentially violating inferences, not how users experience those inferences nor what controls users may want governing their use. We built the Reflective Layer, a visualization tool that surfaces example unstated inferences from users' own ChatGPT histories, and used it in a mixed-methods study with 18 regular ChatGPT users evaluating 215 surfaced inferences from their own conversations. Counterintuitively, participants reacted more strongly with curiosity and interest rather than distress and concern. Discomfort arose mainly when inferences felt misrepresentative of the user or misaligned with expected use. Participants were also markedly less comfortable with advertisers and third-party applications using those inferences than with platform providers. These findings suggest that the acceptability of LLM inferences is governed not only by its content, but by context-sensitive norms around how they are generated, retained within the platform, and transmitted beyond it.
format Preprint
id arxiv_https___arxiv_org_abs_2605_10013
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle When Are LLM Inferences Acceptable? User Reactions and Control Preferences for Inferred Personal Information
Monteiro, Kyzyl
Park, Minjung
Ioffrida, Alexander
Sanna, Angelina
Hao-Ping
Lee
Mireshghallah, Niloofar
Wang, Yang
Das, Sauvik
Human-Computer Interaction
Cryptography and Security
H.5.2; K.4.1; D.4.6
Ask ChatGPT about vacation planning, and it may infer your income. Ask it about medication, and it may infer your medical history. Because such inferences can expose more information than users intend to reveal, prior work argues that they are a defining privacy risk of LLM-based systems. Yet prior work has mostly shown that LLMs can make potentially violating inferences, not how users experience those inferences nor what controls users may want governing their use. We built the Reflective Layer, a visualization tool that surfaces example unstated inferences from users' own ChatGPT histories, and used it in a mixed-methods study with 18 regular ChatGPT users evaluating 215 surfaced inferences from their own conversations. Counterintuitively, participants reacted more strongly with curiosity and interest rather than distress and concern. Discomfort arose mainly when inferences felt misrepresentative of the user or misaligned with expected use. Participants were also markedly less comfortable with advertisers and third-party applications using those inferences than with platform providers. These findings suggest that the acceptability of LLM inferences is governed not only by its content, but by context-sensitive norms around how they are generated, retained within the platform, and transmitted beyond it.
title When Are LLM Inferences Acceptable? User Reactions and Control Preferences for Inferred Personal Information
topic Human-Computer Interaction
Cryptography and Security
H.5.2; K.4.1; D.4.6
url https://arxiv.org/abs/2605.10013