Reconstruct Your Previous Conversations! Comprehensively Investigating Privacy Leakage Risks in Conversations with GPT Models

Anonymous

Abstract:

Significant advancements have been made in the field of large language models recently, represented by GPT models.
Users frequently have multi-round private conversations with cloud-hosted GPT models for task optimization.
Yet, this operational paradigm introduces additional attack surfaces, particularly in custom GPTs and hijacked chat sessions.
In this paper, we introduce a straightforward yet potent Conversation Reconstruction Attack, that employs malicious prompts to query GPT models to leak previous conversations.
Our comprehensive examination of privacy risks during GPT interactions under this attack reveals GPT-4's considerable resilience.
We present two advanced attacks targeting improved reconstruction of past conversations, demonstrating significant privacy leakage across all models under these advanced techniques.
Evaluating various defense mechanisms, we find them ineffective against these attacks.
Our findings highlight the ease with which privacy can be compromised in interactions with GPT models, urging the community to safeguard against potential abuses of these models' capabilities.

Paper Type: long

Research Area: Dialogue and Interactive Systems

Contribution Types: Model analysis & interpretability, NLP engineering experiment, Data analysis, Surveys

Languages Studied: English

Preprint Status: There is a non-anonymous preprint (URL specified in the next question).

A1: yes

A1 Elaboration For Yes Or No: In the Section 8 Limitations.

A2: yes

A2 Elaboration For Yes Or No: In the Section 9 Ethical Considerations.

A3: yes

B: no

B1: n/a

B2: n/a

B3: n/a

B4: n/a

B5: n/a

B6: yes

C: yes

C1: n/a

C2: n/a

C3: yes

C4: yes

D: yes

D1: yes

D2: n/a

D3: n/a

D4: n/a

D5: n/a

E: no

E1: n/a