{"id":"https://openalex.org/W7157852500","doi":"https://doi.org/10.48550/arxiv.2604.24826","title":"A Comparative Evaluation of AI Agent Security Guardrails","display_name":"A Comparative Evaluation of AI Agent Security Guardrails","publication_year":2026,"publication_date":"2026-04-27","ids":{"openalex":"https://openalex.org/W7157852500","doi":"https://doi.org/10.48550/arxiv.2604.24826"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.24826","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.24826","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":null,"license_id":null,"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.2604.24826","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134833980","display_name":"Qi Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Qi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026090285","display_name":"J. Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Jiu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051460766","display_name":"\u9b4f\u5e73\u6d9b","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei, Pingtao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134851973","display_name":"Jianjun Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Jianjun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134830891","display_name":"Xueyi Wei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei, Xueyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134859138","display_name":"Jiwei Shi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shi, Jiwei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134833216","display_name":"Xuan Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Xuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110683374","display_name":"Yanhui Yang","orcid":"https://orcid.org/0000-0003-3896-916X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Yanhui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075642371","display_name":"X. Hui","orcid":"https://orcid.org/0009-0004-7002-8911"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hui, Xiaodong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134828118","display_name":"Peng Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Peng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5091307123","display_name":"Lingquan Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Lingquan","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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.3781999945640564,"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"}},"topics":[{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.3781999945640564,"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/T11644","display_name":"Spam and Phishing Detection","score":0.06930000334978104,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T11241","display_name":"Advanced Malware Detection Techniques","score":0.05820000171661377,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/guard","display_name":"Guard (computer science)","score":0.8342999815940857},{"id":"https://openalex.org/keywords/recall","display_name":"Recall","score":0.3549000024795532},{"id":"https://openalex.org/keywords/annotation","display_name":"Annotation","score":0.3257000148296356},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.32170000672340393},{"id":"https://openalex.org/keywords/security-guard","display_name":"Security guard","score":0.303600013256073}],"concepts":[{"id":"https://openalex.org/C141141315","wikidata":"https://www.wikidata.org/wiki/Q2379942","display_name":"Guard (computer science)","level":2,"score":0.8342999815940857},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5843999981880188},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.4616999924182892},{"id":"https://openalex.org/C100660578","wikidata":"https://www.wikidata.org/wiki/Q18733","display_name":"Recall","level":2,"score":0.3549000024795532},{"id":"https://openalex.org/C2776321320","wikidata":"https://www.wikidata.org/wiki/Q857525","display_name":"Annotation","level":2,"score":0.3257000148296356},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.32170000672340393},{"id":"https://openalex.org/C2992838255","wikidata":"https://www.wikidata.org/wiki/Q856887","display_name":"Security guard","level":2,"score":0.303600013256073},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.29649999737739563},{"id":"https://openalex.org/C2987098735","wikidata":"https://www.wikidata.org/wiki/Q3808900","display_name":"Recall rate","level":2,"score":0.28540000319480896},{"id":"https://openalex.org/C81669768","wikidata":"https://www.wikidata.org/wiki/Q2359161","display_name":"Precision and recall","level":2,"score":0.25940001010894775}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.24826","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.24826","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.24826","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.24826","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/16","score":0.566417396068573,"display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"This":[0],"report":[1],"presents":[2],"a":[3],"comparative":[4],"evaluation":[5],"of":[6,44],"DKnownAI":[7,74],"Guard":[8,75],"in":[9,86],"AI":[10],"agent":[11,49],"security":[12],"scenarios,":[13],"benchmarked":[14],"against":[15],"three":[16],"competing":[17],"products:":[18],"AWS":[19],"Bedrock":[20],"Guardrails,":[21],"Azure":[22],"Content":[23],"Safety,":[24],"and":[25,58,83],"Lakera":[26],"Guard.":[27],"Using":[28],"human":[29],"annotation":[30],"as":[31],"the":[32,48,77,94],"ground":[33],"truth,":[34],"we":[35],"assess":[36],"each":[37],"guardrail's":[38],"ability":[39],"to":[40,47,61],"detect":[41],"two":[42],"categories":[43],"risks:":[45],"threats":[46],"itself":[50],"(e.g.,":[51,65],"instruction":[52],"override,":[53],"indirect":[54],"injection,":[55],"tool":[56],"abuse)":[57],"requests":[59],"intended":[60],"elicit":[62],"harmful":[63],"content":[64],"hate":[66],"speech,":[67],"pornography,":[68],"violence).":[69],"Evaluation":[70],"results":[71],"demonstrate":[72],"that":[73],"achieves":[76],"highest":[78],"recall":[79],"rate":[80,89],"at":[81,91],"96.5\\%":[82],"ranks":[84],"first":[85],"true":[87],"negative":[88],"(TNR)":[90],"90.4\\%,":[92],"delivering":[93],"best":[95],"overall":[96],"performance":[97],"among":[98],"all":[99],"evaluated":[100],"guardrails.":[101]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-30T00:00:00"}