VCT: A Verifiable Transcript System for LLM Conversations
2026-06-22 • Cryptography and Security
Cryptography and Security
AI summaryⓘ
The authors address problems in tracking conversations with large language models (LLMs), which can have complex changes like edits, deletions, and use across devices. They propose Verifiable Conversation Transcript (VCT), a method that uses cryptographic structures to securely record conversation states and changes. Their system ensures records are tamper-evident, consistent, and shareable, even when multiple devices or users interact asynchronously. They also provide protocols to handle deletions without conflicts and detect malicious attempts to alter conversation views. Their tests show the system adds minimal delay and storage overhead, making it practical for real-world forensic uses.
large language modelsdigital forensicstamper-evident logsMerkle treecryptographic hash chainsstate transition protocolnon-repudiationforensic auditinggossip protocoldata integrity
Authors
Ruilin Xing, Feihong Li, Jiayue Liu, Jiali Zheng, Wei Liu, Wanzhi Xie
Abstract
Large language model (LLM) interaction records are increasingly vital in digital forensics and compliance auditing. However, traditional linear tamper-evident logs fail to capture the inherent non-linear evolution of LLM conversations, such as re-prompting based on historical queries, response regeneration, session deletion, multi-device concurrency, and selective sharing. To address this issue, this paper proposes Verifiable Conversation Transcript (VCT), which abstracts complex non-linear LLM semantic operations into account-level authenticated state transitions. VCT constructs a three-tier cryptographic data structure: atomic Q&A pairs form branch-level hash chains, branch tails aggregate into session-level Merkle roots, and all session roots are further aggregated into an account-level Merkle root anchored by joint signatures from both the user and the server. VCT introduces a serialized state transition protocol with deletion barriers to eliminate conflicts between deletion and modification, complemented by a deterministic state-merge protocol to preserve concurrent non-deletion incremental operations. Furthermore, incremental denial checks and a gossip protocol enable asynchronous user devices to autonomously detect view forks caused by malicious servers and generate non-repudiable forensic evidence. Security analysis demonstrates that, under standard cryptographic assumptions, VCT guarantees the integrity, consistency, verifiable shareability, and non-repudiation of account-level conversation records. Evaluation of a Python prototype shows that the cryptographic latency of core operations is within sub-millisecond to low-millisecond ranges. Under a realistic configuration with 21 KB of text, security metadata introduces a negligible storage overhead of only 0.9%, validating the deployment feasibility of VCT for high-stakes forensic review on production-grade LLM platforms.