{"id":"https://openalex.org/W7225200248","doi":"https://doi.org/10.5281/zenodo.23911637","title":"When the Rules Run Out: Can language models find what aircraft certification rules miss? Evidence from 32 years of FAA special conditions","display_name":"When the Rules Run Out: Can language models find what aircraft certification rules miss? Evidence from 32 years of FAA special conditions","publication_year":2026,"publication_date":"2026-11-03","ids":{"openalex":"https://openalex.org/W7225200248","doi":"https://doi.org/10.5281/zenodo.23911637"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.23911637","doi":"https://doi.org/10.5281/zenodo.23911637","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23911637","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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"},"volume":null,"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.5281/zenodo.23911637","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5160460315","display_name":"Gabriel Henrique Antunes Goncalves","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Goncalves, Gabriel Henrique Antunes","raw_affiliation_strings":["Industrial AI Lab"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Industrial AI Lab","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5160460315"],"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/T13643","display_name":"Artificial Intelligence in Law","score":0.49000000953674316,"subfield":{"id":"https://openalex.org/subfields/3308","display_name":"Law"},"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/T13643","display_name":"Artificial Intelligence in Law","score":0.49000000953674316,"subfield":{"id":"https://openalex.org/subfields/3308","display_name":"Law"},"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/T13295","display_name":"Safety Systems Engineering in Autonomy","score":0.1428000032901764,"subfield":{"id":"https://openalex.org/subfields/2213","display_name":"Safety, Risk, Reliability and Quality"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.11869999766349792,"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/aircraft-certification","display_name":"aircraft certification","score":0.7110000252723694},{"id":"https://openalex.org/keywords/language-models","display_name":"language models","score":0.7099999785423279},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.4869999885559082},{"id":"https://openalex.org/keywords/benchmark","display_name":"benchmark","score":0.3370000123977661},{"id":"https://openalex.org/keywords/claude","display_name":"Claude","score":0.6060000061988831},{"id":"https://openalex.org/keywords/airbus-a380","display_name":"Airbus A380","score":0.7129999995231628},{"id":"https://openalex.org/keywords/regulatory-compliance","display_name":"regulatory compliance","score":0.4189999997615814}],"concepts":[{"id":"https://openalex.org/C2777492932","wikidata":"https://www.wikidata.org/wiki/Q1093523","display_name":"Airworthiness","level":3,"score":0.7146000266075134},{"id":"https://openalex.org/C2779151265","wikidata":"https://www.wikidata.org/wiki/Q1156791","display_name":"Copying","level":2,"score":0.6348000168800354},{"id":"https://openalex.org/C46304622","wikidata":"https://www.wikidata.org/wiki/Q374814","display_name":"Certification","level":2,"score":0.6301000118255615},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.5059000253677368},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.46160000562667847},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.44920000433921814},{"id":"https://openalex.org/C42475967","wikidata":"https://www.wikidata.org/wiki/Q194292","display_name":"Operations research","level":1,"score":0.43560001254081726},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4169999957084656},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.39430001378059387},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.3930000066757202},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.3528999984264374},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.3328999876976013},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.32199999690055847},{"id":"https://openalex.org/C2780617661","wikidata":"https://www.wikidata.org/wiki/Q541563","display_name":"Subcategory","level":2,"score":0.31380000710487366},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.2818000018596649},{"id":"https://openalex.org/C54525549","wikidata":"https://www.wikidata.org/wiki/Q2553445","display_name":"Weaving","level":2,"score":0.2612000107765198},{"id":"https://openalex.org/C96865113","wikidata":"https://www.wikidata.org/wiki/Q2946816","display_name":"Certificate","level":2,"score":0.2533000111579895},{"id":"https://openalex.org/C108170787","wikidata":"https://www.wikidata.org/wiki/Q3951828","display_name":"Agency (philosophy)","level":2,"score":0.2522999942302704}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.5281/zenodo.23911637","doi":"https://doi.org/10.5281/zenodo.23911637","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23911637","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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"},{"id":"doi:10.5281/zenodo.23911638","doi":"https://doi.org/10.5281/zenodo.23911638","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23911638","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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"}],"best_oa_location":{"id":"doi:10.5281/zenodo.23911637","doi":"https://doi.org/10.5281/zenodo.23911637","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23911637","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":2,"referenced_works":["https://openalex.org/W7225155924","https://openalex.org/W7225156898"],"related_works":["https://openalex.org/W7167246297","https://openalex.org/W7167720149","https://openalex.org/W7221574776","https://openalex.org/W7159780237","https://openalex.org/W7203717758","https://openalex.org/W4379877205","https://openalex.org/W7223469532","https://openalex.org/W4411120433","https://openalex.org/W7165029580","https://openalex.org/W4283203941"],"abstract_inverted_index":{"Every":[0],"genuinely":[1],"new":[2,202,215],"aircraft":[3],"feature":[4,171],"reaches":[5],"a":[6,60,72,122,142,284],"point":[7],"where":[8,149],"the":[9,29,79,84,103,117,169,181,189,225,239,307,326,333,338,340,349],"airworthiness":[10],"rules":[11],"have":[12],"no":[13,150,206,249,361],"answer.":[14],"The":[15,174],"FAA":[16,80],"then":[17],"writes":[18],"special":[19,48,166],"conditions,":[20],"and":[21,31,52,94,100,196,281,330,337,352],"spotting":[22],"these":[23],"gaps":[24,217],"early":[25],"is":[26,192,273,315,365],"one":[27,108],"of":[28,54,145,238,276,306],"hardest":[30],"most":[32],"consequential":[33],"steps":[34],"in":[35,234,263,294],"certification.":[36],"Language":[37],"models":[38,97,118,213],"are":[39,50,221],"being":[40],"proposed":[41],"as":[42,347],"assistants":[43],"for":[44,168,290,301],"this":[45,133,231,309],"work,":[46],"but":[47,185],"conditions":[49,167],"public":[51],"86%":[53],"them":[55],"repeat":[56],"earlier":[57],"ones,":[58],"so":[59],"model":[61,90,110,160,258],"can":[62],"look":[63,130],"capable":[64],"simply":[65],"by":[66,78,265,356],"recalling":[67],"precedent.":[68,219],"We":[69,86,204],"built":[70],"CertBasis-Bench,":[71],"benchmark":[73,310],"whose":[74],"answers":[75],"were":[76,354],"written":[77],"itself,":[81],"to":[82,120,153],"separate":[83],"two.":[85],"tested":[87],"three":[88],"open-weight":[89,210],"families":[91,351],"(five":[92],"configurations)":[93],"five":[95],"closed":[96,176,212],"from":[98,136,252],"OpenAI":[99],"Anthropic":[101],"on":[102,125,201],"same":[104,170],"314":[105],"items,":[106,148],"plus":[107],"decision":[109],"under":[111],"its":[112,186],"own":[113],"protocol.":[114],"At":[115],"first,":[116],"appeared":[119],"beat":[121],"precedent-based":[123],"baseline":[124],"first-of-kind":[126],"features.":[127,203],"A":[128],"closer":[129],"showed":[131],"that":[132,208,286],"advantage":[134],"comes":[135],"items":[137,336],"published":[138],"before":[139],"2010,":[140],"mostly":[141],"2006":[143],"cluster":[144],"Airbus":[146],"A380":[147],"precedent":[151,190],"existed":[152],"compare":[154],"against.":[155],"From":[156],"2010":[157],"onwards,":[158],"every":[159],"scored":[161],"below":[162],"what":[163],"copying":[164],"later":[165],"would":[172],"give.":[173],"best":[175],"model,":[177],"GPT-6.1":[178],"Sol,":[179],"has":[180],"highest":[182],"overall":[183],"score,":[184],"lead":[187,200],"over":[188],"template":[191],"not":[193,199,366],"statistically":[194],"clear,":[195],"it":[197],"does":[198],"find":[205],"evidence":[207],"current":[209],"or":[211,297,303],"identify":[214],"certification":[216,302],"beyond":[218],"They":[220],"useful":[222],"elsewhere:":[223],"given":[224],"official":[226],"text,":[227],"they":[228],"answer":[229],"\"is":[230],"rule":[232],"still":[233],"force?\"":[235],"correctly":[236],"77\u201389%":[237],"time,":[240],"against":[241,345],"30\u201360%":[242],"without":[243],"it.":[244],"Funding:":[245],"This":[246],"work":[247,305],"received":[248],"specific":[250],"grant":[251],"any":[253],"funding":[254],"agency.":[255],"Compute":[256],"(cloud":[257],"API":[259],"credits)":[260],"was":[261,343],"provided":[262],"kind":[264,308],"Industrial":[266,277],"AI":[267,278,288,299,317,358,364],"Lab.":[268],"Competing":[269],"interests:":[270],"G.":[271],"Gon\u00e7alves":[272],"Vice":[274],"President":[275],"Lab":[279],"(non-profit)":[280],"owns":[282],"SmartHow,":[283],"company":[285],"provides":[287],"solutions":[289],"industry.":[291],"SmartHow":[292,313],"could":[293],"future":[295],"build":[296],"sell":[298],"tools":[300],"compliance":[304],"evaluates.":[311],"No":[312],"product":[314],"evaluated.":[316],"use:":[318],"Claude":[319,322,346],"Code":[320],"(Anthropic;":[321],"Opus":[323],"5.5)":[324],"wrote":[325],"pipeline,":[327],"scoring":[328],"code":[329],"figures,":[331],"drafted":[332],"synthetic":[334],"T1":[335],"paper;":[339],"novelty":[341],"threshold":[342],"calibrated":[344],"coder;":[348],"SC-25":[350],"labels":[353],"produced":[355],"an":[357,367],"agent.":[359],"AI-only;":[360],"human":[362],"verification.":[363],"author.":[368]},"counts_by_year":[],"updated_date":"2026-11-05T12:25:35.769452","created_date":"2026-11-04T00:00:00"}