000 04007naa a22005297i 4500
001 UMUN-OA-20261005-ACC-064
003 local
005 20261005174841.0
006 m o d
007 cr||||||||||||
008 261005s2023 xx o u000 0 eng d
024 7 _a10.1007/s43681-023-00289-2
_2doi
035 _a(local)OpenAlex:W4381572755
040 _alocal
_beng
_clocal
041 0 _aeng
100 0 _aJakob Mökander
_eauthor.
245 1 0 _aAuditing large language models: a three-layered approach /
_cJakob Mökander; Jonas Schuett; Hannah Rose Kirk [and others].
264 1 _a[Place of publication not identified] :
_bSpringer Nature,
_c2023.
300 _a1 online resource (1 article).
336 _atext
_btxt
_2rdacontent
337 _acomputer
_bc
_2rdamedia
338 _aonline resource
_bcr
_2rdacarrier
500 _aPublication type: Journal article.
500 _aLinked open-access copy: published version.
506 0 _aOpen access; no subscription required for the linked open-access version.
_fUnrestricted online access
_2star
520 3 _aAbstract Large language models (LLMs) represent a major advance in artificial intelligence (AI) research. However, the widespread use of LLMs is also coupled with significant ethical and social challenges. Previous research has pointed towards auditing as a promising governance mechanism to help ensure that AI systems are designed and deployed in ways that are ethical, legal, and technically robust. However, existing auditing procedures fail to address the governance challenges posed by LLMs, which display emergent capabilities and are adaptable to a wide range of downstream tasks. In this article, we address that gap by outlining a novel blueprint for how to audit LLMs. Specifically, we propose a three-layered approach, whereby governance audits (of technology providers that design and disseminate LLMs), model audits (of LLMs after pre-training but prior to their release), and application audits (of applications based on LLMs) complement and inform each other. We show how audits, when conducted in a structured and coordinated manner on all three levels, can be a feasible and effective mechanism for identifying and managing some of the ethical and social risks posed by LLMs. However, it is important to remain realistic about what auditing can reasonably be expected to achieve. Therefore, we discuss the limitations not only of our three-layered approach but also of the prospect of auditing LLMs at all. Ultimately, this article seeks to expand the methodological toolkit available to technology providers and policymakers who wish to analyse and evaluate LLMs from technical, ethical, and legal perspectives.
540 _aCreative Commons Attribution (licence family recorded by the source; version not supplied in metadata).
_uhttps://openalex.org/W4381572755
546 _aIn English.
588 0 _aDescription generated from OpenAlex bibliographic metadata retrieved on 5 October 2026; full text linked from the source.
590 _aCourse allocation: B. Sc. Accounting.
650 4 _alarge language models
650 4 _aartificial intelligence in auditing
650 4 _aAI governance
650 4 _aethical risks of AI
650 4 _aAI regulation
653 _aB. Sc. Accounting
653 _aUMUN open access collection
653 _aauditing
655 4 _aJournal articles.
700 0 _aJonas Schuett
_eauthor.
700 0 _aHannah Rose Kirk
_eauthor.
700 0 _aLuciano Floridi
_eauthor.
773 0 _tAI and Ethics
_gVol. 4, no. 4, (2023), pages/article 1085-1115
_x2730-5953
856 4 0 _uhttps://link.springer.com/content/pdf/10.1007/s43681-023-00289-2.pdf
_yRead open-access publication
_zOpen access. Access and licence reported by the source; this host could not be independently checked by automation.
_70
884 _aCourse selection and automated MARC21 enrichment from OpenAlex metadata.
_g20261005
_khttps://openalex.org/W4381572755
_qlocal
942 _n0
999 _c816
_d816