Uses WildChat
Trust No Bot: Discovering Personal Disclosures in Human-LLM Conversations in the Wild
COLM 2024 · First arXiv submission
Studies the tasks and contexts in which people disclose personal or sensitive information to chatbots. It combines automated PII detection with task and sensitive-topic taxonomies, releasing annotations that support analyses beyond conventional identifiers.
Data used
A 5,000-conversation English sample from the one-million-conversation WildChat corpus, limited to one conversation per hashed IP with additional prefix and length filters.
Findings
Sensitive disclosures in the analyzed sample include preferences, personal habits, and application-related information that conventional PII detectors miss.
Section 2: sampling. Sections 3–4: detected identifiers and sensitive topics. Read the source.
Dates and versions
First arXiv version. ArXiv version history.
2024 arXiv preprint; also published at COLM 2024.
Cite this paper
Niloofar Mireshghallah, Maria Antoniak, Yash More, Yejin Choi, Golnoosh Farnadi. 2024. Trust No Bot: Discovering Personal Disclosures in Human-LLM Conversations in the Wild. arXiv preprint.
Show BibTeX
@misc{mireshghallah2024trustnobot,
title = {{Trust No Bot: Discovering Personal Disclosures in Human-LLM Conversations in the Wild}},
author = {Mireshghallah, Niloofar and Antoniak, Maria and More, Yash and Choi, Yejin and Farnadi, Golnoosh},
year = {2024},
doi = {10.48550/arXiv.2407.11438},
eprint = {2407.11438},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2407.11438}
}