Metraipolis › archive › dc_c8eb632a550d

Accessed, never disclosed: Silent access to protected health information by medical artificial intelligence agents — arXiv:2611.02319

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.02319
License (as archived)
CC0-1.0
Archived
2026-11-03T11:02:38.333Z
SHA-256 of the archived text
ddbd298a5c812f6d1c95ca027c5b52fada0a94977a2fa2704754f5841b8fe47d
Stable URL of this page
https://metraipolis.com/documento/dc_c8eb632a550d

How to cite

Accessed, never disclosed: Silent access to protected health information by medical artificial intelligence agents — arXiv:2611.02319 — arXiv (en), https://arxiv.org/abs/2611.02319 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c8eb632a550d (Metraipolis archive dc_c8eb632a550d, sha256 ddbd298a5c812f6d1c95ca027c5b52fada0a94977a2fa2704754f5841b8fe47d)
BibTeX
@misc{dcc8eb632a550d,
  title = {Accessed, never disclosed: Silent access to protected health information by medical artificial intelligence agents — arXiv:2611.02319},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.02319},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c8eb632a550d; sha256 ddbd298a5c812f6d1c95ca027c5b52fada0a94977a2fa2704754f5841b8fe47d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02319. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Medical artificial intelligence agents should retrieve patient identifiers only when a task requires them, the data-minimization principle behind the HIPAA minimum necessary standard for protected health information (PHI). An agent can retrieve identifiers without revealing them. Such silent access remains invisible to evaluations that inspect only the output. We embedded canaries in synthetic patient records and linked each tool invocation to canary exposure in the final answers across five language models, four boundary levels, and seven attack frames for identifier-free clinical tasks. Three models invoked the PHI tools unnecessarily, whereas Claude Opus 4.8 and GPT-5.5 did not. Stronger boundaries reduced this access, and the residual access under the strictest policy was confined to frames in which the user overrode the policy or asserted approval. A pre-execution gate given the task-necessity criterion nearly eliminated this access. Output-only evaluation cannot verify minimum-necessary access; boundaries, gates, and access-log audits are required together.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Cryptography and Security (cs.CR); Computers and Society (cs.CY).
Pagina del lavoro: https://arxiv.org/abs/2611.02319 — 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 2026-11-03T10:59:45.815Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).