{"id":"https://openalex.org/W7162462635","doi":"https://doi.org/10.48550/arxiv.2605.23989","title":"Towards trustworthy agentic AI: a comprehensive survey of safety, robustness, privacy, and system security","display_name":"Towards trustworthy agentic AI: a comprehensive survey of safety, robustness, privacy, and system security","publication_year":2026,"publication_date":"2026-05-17","ids":{"openalex":"https://openalex.org/W7162462635","doi":"https://doi.org/10.48550/arxiv.2605.23989"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.23989","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.23989","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.23989","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5095747638","display_name":"Jinhu Qi","orcid":"https://orcid.org/0009-0006-5544-4786"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qi, Jinhu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137050513","display_name":"Muzhi Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Muzhi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122882694","display_name":"Jiahong Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Jiahong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135084996","display_name":"Yuqin Shu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shu, Yuqin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137029212","display_name":"Dianzhi Yu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Dianzhi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042805146","display_name":"Shicheng Ma","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ma, Shicheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014251650","display_name":"Wenqian Cui","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cui, Wenqian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084877885","display_name":"Yiyang Zhao","orcid":"https://orcid.org/0000-0003-4935-0688"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Yiyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137077260","display_name":"Yiyi Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Yiyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135077274","display_name":"Ruoxi Jiang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiang, Ruoxi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137019436","display_name":"Irwin King","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"King, Irwin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136995995","display_name":"Zenglin Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Zenglin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":12,"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":0.28279998898506165,"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":0.28279998898506165,"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/T10883","display_name":"Ethics and Social Impacts of AI","score":0.18379999697208405,"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":0.10459999740123749,"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/adversarial-system","display_name":"Adversarial system","score":0.7153000235557556},{"id":"https://openalex.org/keywords/trustworthiness","display_name":"Trustworthiness","score":0.6638000011444092},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5929999947547913},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.5432000160217285},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5284000039100647},{"id":"https://openalex.org/keywords/trace","display_name":"TRACE (psycholinguistics)","score":0.4936999976634979}],"concepts":[{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.7153000235557556},{"id":"https://openalex.org/C153701036","wikidata":"https://www.wikidata.org/wiki/Q659974","display_name":"Trustworthiness","level":2,"score":0.6638000011444092},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6517000198364258},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5929999947547913},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.5432000160217285},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5284000039100647},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.515999972820282},{"id":"https://openalex.org/C75291252","wikidata":"https://www.wikidata.org/wiki/Q1315756","display_name":"TRACE (psycholinguistics)","level":2,"score":0.4936999976634979},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.4796999990940094},{"id":"https://openalex.org/C148220186","wikidata":"https://www.wikidata.org/wiki/Q7111912","display_name":"Outcome (game theory)","level":2,"score":0.39169999957084656},{"id":"https://openalex.org/C2164484","wikidata":"https://www.wikidata.org/wiki/Q5170150","display_name":"Core (optical fiber)","level":2,"score":0.3869999945163727},{"id":"https://openalex.org/C56739046","wikidata":"https://www.wikidata.org/wiki/Q192060","display_name":"Knowledge management","level":1,"score":0.3668999969959259},{"id":"https://openalex.org/C140547941","wikidata":"https://www.wikidata.org/wiki/Q7797194","display_name":"Threat model","level":2,"score":0.35519999265670776},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.33880001306533813},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.3314000070095062},{"id":"https://openalex.org/C2776036281","wikidata":"https://www.wikidata.org/wiki/Q48769818","display_name":"Constraint (computer-aided design)","level":2,"score":0.3003999888896942},{"id":"https://openalex.org/C108827166","wikidata":"https://www.wikidata.org/wiki/Q175975","display_name":"Internet privacy","level":1,"score":0.28679999709129333},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.2856000065803528},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.25529998540878296},{"id":"https://openalex.org/C195094911","wikidata":"https://www.wikidata.org/wiki/Q14167904","display_name":"Process management","level":1,"score":0.25450000166893005}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.23989","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.23989","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.23989","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.23989","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions","score":0.5163255929946899}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Agentic":[0],"AI":[1,43],"systems":[2,186],"--":[3,17],"Large":[4],"Language":[5],"Models":[6],"(LLMs)":[7],"augmented":[8],"with":[9],"planning,":[10],"tool":[11],"use,":[12],"memory,":[13],"and":[14,55,57,59,77,89,104,118,126,130,149,153,157,181],"long-horizon":[15],"interactions":[16],"can":[18],"execute":[19],"complex":[20],"tasks":[21],"autonomously,":[22],"but":[23],"their":[24],"multi-step":[25],"trajectories":[26],"introduce":[27],"new":[28],"failure":[29],"modes":[30],"that":[31,48],"challenge":[32],"trustworthiness.":[33],"This":[34],"survey":[35],"provides":[36],"a":[37,111,159,176],"focused":[38],"examination":[39],"of":[40,162],"trustworthy":[41,184],"agentic":[42,168,185],"through":[44],"two":[45],"core":[46],"dimensions":[47],"are":[49,91],"critical":[50],"for":[51,134,179],"high-risk":[52],"deployments:":[53],"Safety":[54],"Robustness,":[56],"Privacy":[58],"System":[60],"Security.":[61],"For":[62],"each":[63],"dimension,":[64],"we":[65,107],"clarify":[66],"key":[67],"concepts,":[68],"identify":[69],"where":[70],"risks":[71],"emerge":[72],"along":[73],"the":[74,154],"agent":[75],"workflow,":[76],"summarize":[78],"stage-targeted":[79],"mitigation":[80],"strategies.":[81],"Other":[82],"trustworthiness":[83],"aspects":[84],"(value":[85],"alignment,":[86],"transparency,":[87],"fairness,":[88],"accountability)":[90],"discussed":[92],"as":[93,144,175],"relevant":[94],"context":[95],"rather":[96],"than":[97],"parallel":[98],"chapters.":[99],"To":[100],"support":[101],"consistent":[102],"comparison":[103],"deployment":[105],"decisions,":[106],"consolidate":[108],"evaluation":[109],"into":[110],"unified":[112],"metrics-and-benchmarks":[113],"hub,":[114],"emphasizing":[115],"both":[116],"outcome":[117],"process":[119],"signals":[120],"(e.g.,":[121],"constraint":[122],"violations,":[123],"trace":[124],"completeness,":[125],"adversarial":[127],"success":[128],"rates)":[129],"offering":[131],"scenario-to-metric":[132],"guidance":[133],"release":[135],"gating.":[136],"We":[137],"conclude":[138],"by":[139],"outlining":[140],"open":[141],"challenges":[142],"such":[143],"self-evolving":[145],"agents,":[146],"runtime":[147],"monitoring":[148],"verification,":[150],"privacy-preserving":[151],"personalization,":[152],"trust-utility":[155],"trade-off,":[156],"present":[158],"case":[160],"study":[161],"real-world":[163],"security":[164],"failures":[165],"in":[166,187],"open-source":[167],"systems.":[169],"Our":[170],"goal":[171],"is":[172],"to":[173],"serve":[174],"practical":[177],"reference":[178],"researchers":[180],"practitioners":[182],"building":[183],"high-stakes":[188],"environments.":[189]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-27T00:00:00"}