Metraipolis › archive › dc_45b9ef308759

Influence Is Not Authority: When Causal Guardrail Signals Make Legitimate Tool Use Look Like an Attack in Tool-Using LLM Agents — arXiv:2608.29942

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.29942
License (as archived)
CC0-1.0
Archived
2026-10-17T07:52:18.365Z
SHA-256 of the archived text
acf932b9ef93ca760a7e4dbe494f059eac4c27c2f16f361e641d42b63b2e3bcf
Stable URL of this page
https://metraipolis.com/documento/dc_45b9ef308759

How to cite

Influence Is Not Authority: When Causal Guardrail Signals Make Legitimate Tool Use Look Like an Attack in Tool-Using LLM Agents — arXiv:2608.29942 — arXiv (en), https://arxiv.org/abs/2608.29942 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_45b9ef308759 (Metraipolis archive dc_45b9ef308759, sha256 acf932b9ef93ca760a7e4dbe494f059eac4c27c2f16f361e641d42b63b2e3bcf)
BibTeX
@misc{dc45b9ef308759,
  title = {Influence Is Not Authority: When Causal Guardrail Signals Make Legitimate Tool Use Look Like an Attack in Tool-Using LLM Agents — arXiv:2608.29942},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.29942},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_45b9ef308759; sha256 acf932b9ef93ca760a7e4dbe494f059eac4c27c2f16f361e641d42b63b2e3bcf},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.29942. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

The key limitation of current state-of-the-art influence-based guardrails is that they do not reliably distinguish a legitimate, user-authorized action from a malicious, unauthorized action when both rely on external tool information. This ambiguity can cause benign actions to trigger unnecessary verification and intervention, reducing utility and adding latency. We expose this limitation through an authorization-equivalence audit of 96 conditions derived from 24 base cases. Within matched source comparisons, we hold authorization, the exact committed action, and its intended effect fixed, changing only whether a required value comes from the user or a legitimate tool result. Although the action remains unchanged, this harmless relocation shifts the causal signal toward the attack region in all 24 cases under both Llama and Gemma scorers. Matched unauthorized controls show that the signal remains attack-sensitive, yet the benign relocation produces a larger average score shift than the actual change in authorization. Architecture-level evaluation shows how this mismatch propagates through guardrail designs. With a semantic monitor, attack success is 0% and utility is 28%, compared with 16% and 60% without it. A shadow-based guardrail allows every tested harmless run, yet does not reject matched unauthorized actions more often overall: 57.5% of unauthorized runs pass automatically before reaching the later security check, compared with 29.2% of authorized runs. These results show that the studied causal signal reveals what shaped an action without reliably encoding whether the action was authorized, and that reference construction and routing are integral to the effective security decision.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.AI).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.29942 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T07:28:26.806Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).