# The Impact of a Chatbot's Ephemerality-Framing on Self-Disclosure Perceptions

arXiv: https://arxiv.org/abs/2505.20464 · HTML: https://mlsanity.com/paper/2505.20464/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 109 references that are in the map.
- First posted on arXiv 26 May 2025.

## Builds on (most cited first, 13 of 13)

- [Extracting Training Data from Large Language Models](https://mlsanity.com/paper/2012.07805/) (2020)
- [What Does it Mean for a Language Model to Preserve Privacy?](https://mlsanity.com/paper/2202.05520/) (2022)
- ["It's a Fair Game", or Is It? Examining How Users Navigate Disclosure Risks and Benefits When Using LLM-Based Conversational Agents](https://mlsanity.com/paper/2309.11653/) (2023)
- [Conceptual Metaphors Impact Perceptions of Human-AI Collaboration](https://mlsanity.com/paper/2008.02311/) (2020)
- [ChaCha: Leveraging Large Language Models to Prompt Children to Share Their Emotions about Personal Events](https://mlsanity.com/paper/2309.12244/) (2023)
- [Protecting Users From Themselves: Safeguarding Contextual Privacy in Interactions with Conversational Agents](https://mlsanity.com/paper/2502.18509/) (2025)
- [ComPeer: A Generative Conversational Agent for Proactive Peer Support](https://mlsanity.com/paper/2407.18064/) (2024)
- [Chatbots for Data Collection in Surveys: A Comparison of Four Theory-Based Interview Probes](https://mlsanity.com/paper/2503.08582/) (2025)
- [Comparing How a Chatbot References User Utterances from Previous Chatting Sessions: An Investigation of Users' Privacy Concerns and Perceptions](https://mlsanity.com/paper/2308.04879/) (2023)
- [Users' Mental Models of Generative AI Chatbot Ecosystems](https://mlsanity.com/paper/2501.19211/) (2025)
- [Submitting surveys via a conversational interface: an evaluation of user acceptance and approach effectiveness](https://mlsanity.com/paper/2003.02537/) (2020)
- [AI-Enabled Conversational Journaling for Advancing Parkinson's Disease Symptom Tracking](https://mlsanity.com/paper/2503.03532/) (2025)
- [What Types of Questions Require Conversation to Answer? A Case Study of AskReddit Questions](https://mlsanity.com/paper/2303.17710/) (2023)

## Built on by (most cited first, 6 of 6)

- [Does My Chatbot Have an Agenda? Understanding Human and AI Agency in Human-Human-like Chatbot Interaction](https://mlsanity.com/paper/2601.22452/) (2026)
- [AlienLM: Alienization of Language for API-Boundary Privacy in Black-Box LLMs](https://mlsanity.com/paper/2601.22710/) (2026)
- [User Perceptions vs. Proxy LLM Judges: Privacy and Helpfulness in LLM Responses to Privacy-Sensitive Scenarios](https://mlsanity.com/paper/2510.20721/) (2025)
- [Chaplains' Reflections on the Design and Usage of AI for Conversational Care](https://mlsanity.com/paper/2602.04017/) (2026)
- [Theory of Mind and Self-Disclosure to CUIs](https://mlsanity.com/paper/2507.10773/) (2025)
- [Gricea: An Open Science Platform for Conversational AI Research](https://mlsanity.com/paper/2609.22039/) (2026)

## Abstract (arXiv)

Self-disclosure, the sharing of one's thoughts and feelings, is affected by the perceived relationship between individuals. While chatbots are increasingly used for self-disclosure, the impact of a chatbot's framing on users' self-disclosure remains under-explored. We investigated how a chatbot's description of its relationship with users, particularly in terms of ephemerality, affects self-disclosure. Specifically, we compared a Familiar chatbot, presenting itself as a companion remembering past interactions, with a Stranger chatbot, presenting itself as a new, unacquainted entity in each conversation. In a mixed factorial design, participants engaged with either the Familiar or Stranger chatbot in two sessions across two days, with one conversation focusing on Emotional- and another Factual-disclosure. When Emotional-disclosure was sought in the first chatting session, Stranger-condition participants felt more comfortable self-disclosing. However, when Factual-disclosure was sought first, these differences were replaced by more enjoyment among Familiar-condition participants. Qualitative findings showed Stranger afforded anonymity and reduced judgement, whereas Familiar sometimes felt intrusive unless rapport was built via low-risk Factual-disclosure.