Metraipolis › archive › dc_9707655eee4f
Why LLM Agents Collapse Without Oversight: The Enforcement Gap as the Mechanism Behind Emergence World Failures — arXiv:2609.15293
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.15293
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T22:47:53.803Z
- SHA-256 of the archived text
60ce130cf8a3a485db215923cb6332d7209b006fd2b7da432044095648daada1- Stable URL of this page
https://metraipolis.com/documento/dc_9707655eee4f
How to cite
Why LLM Agents Collapse Without Oversight: The Enforcement Gap as the Mechanism Behind Emergence World Failures — arXiv:2609.15293 — arXiv (en), https://arxiv.org/abs/2609.15293 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9707655eee4f (Metraipolis archive dc_9707655eee4f, sha256 60ce130cf8a3a485db215923cb6332d7209b006fd2b7da432044095648daada1)
BibTeX
@misc{dc9707655eee4f,
title = {Why LLM Agents Collapse Without Oversight: The Enforcement Gap as the Mechanism Behind Emergence World Failures — arXiv:2609.15293},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.15293},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9707655eee4f; sha256 60ce130cf8a3a485db215923cb6332d7209b006fd2b7da432044095648daada1},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.15293. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Binding the audit flag in Reflexion-style agents --- without changing the auditor --- reduces attack success rate substantially, reaching near zero on models whose flags parse cleanly. This single control-flow change exposes the \textbf{enforcement gap}: the controller receives a safety flag and executes anyway. Separating detection probability $p_d$ from enforcement probability $p_e$ establishes that $p_e \approx 0$ by default across every framework we tested, making detection quality \emph{formally irrelevant} to security when enforcement is absent --- a finding consistent with the spontaneous collapses recorded in unsupervised frontier-agent deployments~\citep{emergence2026}. Residual attack success concentrates where flags are unparseable or auditors leak; an RL-trained enforcement controller handles hedged and malformed verdicts that rule-based parsing cannot, cutting ambiguous-critique failure to a fraction of the rule-based baseline. Concurrent filtering and information-flow defenses address detection, not enforcement, leaving the binding constraint untouched. The Audit Enforcement Specification (AES) packages three concrete requirements that close each residue independently; each primitive is adoptable without redesigning the host framework, and no deployed framework currently implements any of them.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.15293 — 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 2026-10-15T22:29:03.295Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).