Metraipolis › archive › dc_0e7d8db93039

Local Allow, Global Deny: Why Per-Action Authorization Does Not Compose in LLM Agents — arXiv:2610.27620

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.27620
License (as archived)
CC0-1.0
Archived
2026-10-21T07:07:57.216Z
SHA-256 of the archived text
8c116b03898ee86f99145766928323bb3323ccc7df27c8952288f60e80010d9d
Stable URL of this page
https://metraipolis.com/documento/dc_0e7d8db93039

How to cite

Local Allow, Global Deny: Why Per-Action Authorization Does Not Compose in LLM Agents — arXiv:2610.27620 — arXiv (en), https://arxiv.org/abs/2610.27620 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0e7d8db93039 (Metraipolis archive dc_0e7d8db93039, sha256 8c116b03898ee86f99145766928323bb3323ccc7df27c8952288f60e80010d9d)
BibTeX
@misc{dc0e7d8db93039,
  title = {Local Allow, Global Deny: Why Per-Action Authorization Does Not Compose in LLM Agents — arXiv:2610.27620},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.27620},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0e7d8db93039; sha256 8c116b03898ee86f99145766928323bb3323ccc7df27c8952288f60e80010d9d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.27620. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Autonomous LLM agents act through tools under per-action authorization: a local guard screens each tool call in isolation. We show that local authorization does not compose into global authorization. A compositional authorization failure (CAF) is a locally authorized execution, or a pair of locally authorized executions, that violates a global policy spanning intent, purpose, provenance, consent, and delegation. We characterize CAFs along two axes, expressiveness (single-trace safety versus relational k-safety) and observability (a ladder of monitor profiles). Our central result is an impossibility: relational purpose violations form a 2-safety hyperproperty with no per-trace locus of blame, so a per-action monitor that never raises a false alarm detects none of them, independent of the underlying model or decision procedure; they become visible only under cross-execution monitoring, which none of the six guardrails we survey implements. We make this constructive with AL-SYNTH, a SAT/SMT synthesizer (using no language model) that emits minimal CAF witnesses, each checked against an exhaustive oracle. The empirical signature is sharp: across nine LLM judges the per-action view approves the synthesized witness almost without exception, blocking it at a rate comparable to a benign control, yet the same models flag it when shown the two executions together. Per-action authorization is therefore structurally insufficient for relational policy conjuncts not reducible to single-trace safety: to bound them, a monitor must compare executions or accept false alarms.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO).
Pagina del lavoro: https://arxiv.org/abs/2610.27620 — 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 2026-10-21T07:05:43.585Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).