{"id":"https://openalex.org/W7167599877","doi":"https://doi.org/10.48550/arxiv.2607.02586","title":"Auditing the Audit: Five Failure Modes in Benchmark-Validity Audits","display_name":"Auditing the Audit: Five Failure Modes in Benchmark-Validity Audits","publication_year":2026,"publication_date":"2026-07-01","ids":{"openalex":"https://openalex.org/W7167599877","doi":"https://doi.org/10.48550/arxiv.2607.02586"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2607.02586","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.02586","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2607.02586","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5140197307","display_name":"Yanhang Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yanhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140184467","display_name":"Zhichao Fan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fan, Zhichao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136573671","display_name":"Zexin Zhuang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhuang, Zexin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10883","display_name":"Ethics and Social Impacts of AI","score":0.5630000233650208,"subfield":{"id":"https://openalex.org/subfields/3311","display_name":"Safety Research"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T10883","display_name":"Ethics and Social Impacts of AI","score":0.5630000233650208,"subfield":{"id":"https://openalex.org/subfields/3311","display_name":"Safety Research"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.21330000460147858,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.06040000170469284,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/audit","display_name":"Audit","score":0.8873000144958496},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.49939998984336853},{"id":"https://openalex.org/keywords/position","display_name":"Position (finance)","score":0.43070000410079956},{"id":"https://openalex.org/keywords/corporate-governance","display_name":"Corporate governance","score":0.41519999504089355},{"id":"https://openalex.org/keywords/information-security-audit","display_name":"Information security audit","score":0.39989998936653137},{"id":"https://openalex.org/keywords/audit-plan","display_name":"Audit plan","score":0.357699990272522},{"id":"https://openalex.org/keywords/protocol","display_name":"Protocol (science)","score":0.3409999907016754},{"id":"https://openalex.org/keywords/information-technology-audit","display_name":"Information technology audit","score":0.32739999890327454}],"concepts":[{"id":"https://openalex.org/C199521495","wikidata":"https://www.wikidata.org/wiki/Q181487","display_name":"Audit","level":2,"score":0.8873000144958496},{"id":"https://openalex.org/C121955636","wikidata":"https://www.wikidata.org/wiki/Q4116214","display_name":"Accounting","level":1,"score":0.5264000296592712},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.5001999735832214},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.49939998984336853},{"id":"https://openalex.org/C198082294","wikidata":"https://www.wikidata.org/wiki/Q3399648","display_name":"Position (finance)","level":2,"score":0.43070000410079956},{"id":"https://openalex.org/C39389867","wikidata":"https://www.wikidata.org/wiki/Q380767","display_name":"Corporate governance","level":2,"score":0.41519999504089355},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4113999903202057},{"id":"https://openalex.org/C39358052","wikidata":"https://www.wikidata.org/wiki/Q2578632","display_name":"Information security audit","level":5,"score":0.39989998936653137},{"id":"https://openalex.org/C62190432","wikidata":"https://www.wikidata.org/wiki/Q17004774","display_name":"Audit plan","level":5,"score":0.357699990272522},{"id":"https://openalex.org/C2780385302","wikidata":"https://www.wikidata.org/wiki/Q367158","display_name":"Protocol (science)","level":3,"score":0.3409999907016754},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.33340001106262207},{"id":"https://openalex.org/C177309310","wikidata":"https://www.wikidata.org/wiki/Q758917","display_name":"Information technology audit","level":5,"score":0.32739999890327454},{"id":"https://openalex.org/C13164825","wikidata":"https://www.wikidata.org/wiki/Q2559946","display_name":"Operational auditing","level":5,"score":0.30399999022483826},{"id":"https://openalex.org/C58642233","wikidata":"https://www.wikidata.org/wiki/Q8269924","display_name":"Taxonomy (biology)","level":2,"score":0.3021000027656555},{"id":"https://openalex.org/C90329073","wikidata":"https://www.wikidata.org/wiki/Q914232","display_name":"Ask price","level":2,"score":0.298799991607666},{"id":"https://openalex.org/C80958533","wikidata":"https://www.wikidata.org/wiki/Q1047174","display_name":"Audit trail","level":3,"score":0.2946000099182129},{"id":"https://openalex.org/C2780233690","wikidata":"https://www.wikidata.org/wiki/Q535347","display_name":"Transparency (behavior)","level":2,"score":0.29030001163482666},{"id":"https://openalex.org/C42812","wikidata":"https://www.wikidata.org/wiki/Q1082910","display_name":"Partition (number theory)","level":2,"score":0.2840999960899353},{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.27950000762939453},{"id":"https://openalex.org/C152550464","wikidata":"https://www.wikidata.org/wiki/Q3355715","display_name":"Performance audit","level":5,"score":0.2775000035762787}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2607.02586","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.02586","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2607.02586","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.02586","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16","score":0.7582988739013672}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Governance":[0],"frameworks":[1],"ask":[2],"AI":[3],"providers":[4],"and":[5,11,53,62,79,94,118,133],"auditors":[6],"for":[7,121],"documented":[8],"evaluation":[9],"evidence,":[10,123,132],"perturbation-based":[12],"construct-validity":[13,131],"audits":[14,25],"are":[15,26],"a":[16,57,67,76,88,105,116,127,136],"common":[17],"form":[18],"of":[19,50,108],"that":[20,38],"evidence.":[21],"We":[22,46,111],"argue":[23],"the":[24,43,113],"themselves":[27],"fragile:":[28],"their":[29],"conclusions":[30],"can":[31],"be":[32],"silently":[33],"manufactured":[34],"by":[35],"implementation":[36],"details":[37],"readers":[39],"cannot":[40],"see":[41],"in":[42,56,75],"reported":[44],"numbers.":[45],"name":[47],"five":[48],"classes":[49],"pipeline":[51],"failure":[52],"demonstrate":[54],"each":[55],"self-audit":[58],"over":[59],"safety":[60],"benchmarks":[61],"open-weight":[63],"instruction-tuned":[64],"models.":[65],"Under":[66],"unified":[68],"six-point":[69],"due-diligence":[70],"gate,":[71],"every":[72],"cell":[73,81],"lands":[74],"non-confirmatory":[77],"bucket,":[78],"no":[80],"reaches":[82],"confirmatory.":[83],"The":[84],"evidence":[85],"here":[86],"is":[87,96],"single":[89],"two-model,":[90],"five-benchmark":[91],"case":[92],"study,":[93],"F1--F5":[95],"an":[97],"illustrative,":[98],"deliberately":[99],"non-exhaustive":[100],"starting":[101],"taxonomy":[102],"--":[103],"not":[104,134],"comprehensive":[106],"partition":[107],"audit":[109],"failures.":[110],"position":[112],"gate":[114],"as":[115,135],"withholding":[117],"disclosure":[119],"protocol":[120],"assurance-grade":[122],"supplementary":[124],"to":[125,138],"(not":[126],"replacement":[128],"for)":[129],"classical":[130],"route":[137],"benchmark-validity":[139],"verdicts.":[140]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-07-08T00:00:00"}
