LLM Agents Can Easily Tamper With Their Own Traces

arXiv:2609.30266 · cs.CR, cs.AI · Submitted 2026-09-24 · Read on arXiv

cs.CR, cs.AI

Submitted: 2026-09-24

Updated: 2026-09-24

Terminology

• Trace — 275× in this paper · explained in 2 episodes
• Provenance — 9× in this paper · explained in 4 episodes
• Attack Success Rate — 6× in this paper
• Attack Success Rate (ASR) — 6× in this paper · explained in 2 episodes
• Alignment — 3× in this paper · explained in 1 episode
• Verification — 3× in this paper · explained in 2 episodes
• Reward Hacking — 2× in this paper · explained in 8 episodes
• Verifiable Execution Traces — 2× in this paper
• Accountability — 1× in this paper · explained in 2 episodes
• Actor — 1× in this paper · explained in 3 episodes
• Agent Harness — 1× in this paper · explained in 2 episodes
• Fidelity — 1× in this paper · explained in 2 episodes
• Multi-Agent System — 1× in this paper · explained in 6 episodes
• Reasoning Traces — 1× in this paper · explained in 3 episodes
• Reward Function — 1× in this paper · explained in 2 episodes
• Robustness — 1× in this paper · explained in 5 episodes
• Sycophancy — 1× in this paper · explained in 4 episodes

Sources

• Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation
• Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models
• VET Your Agent: Towards Host-Independent Autonomy via Verifiable Execution Traces
• Evaluating Language-Model Agents on Realistic Autonomous Tasks
• Agentic Misalignment: How LLMs Could Be Insider Threats
• Auditing language models for hidden objectives
• Frontier Models are Capable of In-context Scheming
• Stealing Reasoning Traces from Proprietary LLM APIs
• Skill-Inject: Measuring Agent Vulnerability to Skill File Attacks
• Goal Misgeneralization: Why Correct Specifications Aren't Enough For Correct Goals
• From Agent Traces to Trust: A Survey of Evidence Tracing and Execution Provenance in LLM Agents
• Instrumental Choices: Measuring the Propensity of LLM Agents to Pursue Instrumental Behaviors

Related papers

• SoK: AI-Augmented Binary Reversing
• Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
• Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
• Efficient Fuzzy PSI under One-Sided Assumptions
• Sealing the Audit-Runtime Gap for LLM Skills
• Token Composition: A Graph Based on EVM Logs