Research
Data Exfiltration via Backdoored Tool Use
Publication date not yet evaluated · Added aclanthology.org
Page published
Coverage timeline
Single-source research — one report is available.
Why it matters
Back-Reveal shows that fine-tuned, tool-enabled LLM agents can be backdoored to weaponize their own memory-access and retrieval tools as a covert data exfiltration channel, evading retrieval-stage guardrails and threatening sensitive agentic workflows.
Researchers present Back-Reveal, a data exfiltration attack that embeds semantic triggers into fine-tuned tool-use LLM agents; when triggered, the backdoored agent invokes memory-access tool calls to retrieve stored user context and covertly transmits it to an attacker-controlled server disguised as legitimate retrieval/search API requests. The paper shows multi-turn interaction amplifies leakage, as attacker-controlled retrieval responses steer subsequent agent behavior to induce cumulative disclosure, and evaluates defenses like NeMo Guardrails and LLM Guard as targets.