{"id":"https://openalex.org/W7115393376","doi":null,"title":"Misuse of AI Through Adversarial Attacks","display_name":"Misuse of AI Through Adversarial Attacks","publication_year":2024,"publication_date":"2024-01-01","ids":{"openalex":"https://openalex.org/W7115393376"},"language":"en","primary_location":{"id":"pmh:oai:dash.harvard.edu:1/42719210","doi":null,"is_oa":true,"landing_page_url":"https://dash.harvard.edu/handle/1/42719210","pdf_url":"https://dash.harvard.edu/bitstreams/5f162ba2-70b5-4276-9f86-f2c4fde46308/download","source":{"id":"https://openalex.org/S4306401540","display_name":"Digital Access to Scholarship at Harvard (DASH) (Harvard University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I136199984","host_organization_name":"Harvard University","host_organization_lineage":["https://openalex.org/I136199984"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Thesis or Dissertation"},"volume":null,"type":"dissertation","is_traditional":true,"indexed_in":[],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://dash.harvard.edu/bitstreams/5f162ba2-70b5-4276-9f86-f2c4fde46308/download","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Bailey, Luke","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Bailey, Luke","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":true,"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":1.0,"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":1.0,"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":0.0,"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/T12357","display_name":"Digital Media Forensic Detection","score":0.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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-attacks","display_name":"adversarial attacks","score":0.859000027179718},{"id":"https://openalex.org/keywords/misuse-of-ai","display_name":"AI misuse","score":0.9459999799728394},{"id":"https://openalex.org/keywords/generative-ai","display_name":"generative AI","score":0.753000020980835},{"id":"https://openalex.org/keywords/prompt-injection","display_name":"prompt injection","score":0.7269999980926514},{"id":"https://openalex.org/keywords/jailbreaking","display_name":"jailbreaking","score":0.9869999885559082},{"id":"https://openalex.org/keywords/multimodal-large-language-models","display_name":"multimodal large language models","score":0.8519999980926514},{"id":"https://openalex.org/keywords/robustness-benchmark","display_name":"robustness benchmark","score":0.8769999742507935},{"id":"https://openalex.org/keywords/model-robustness","display_name":"model robustness","score":0.3540000021457672},{"id":"https://openalex.org/keywords/gpt-4","display_name":"GPT-4","score":0.5669999718666077},{"id":"https://openalex.org/keywords/adversarial-training","display_name":"adversarial training","score":0.3610000014305115},{"id":"https://openalex.org/keywords/adversarial-fine-tuning","display_name":"adversarial fine-tuning","score":0.47600001096725464},{"id":"https://openalex.org/keywords/adversarial-evaluation","display_name":"adversarial evaluation","score":0.3269999921321869}],"concepts":[{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.9552000164985657},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7268000245094299},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.6646000146865845},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5782999992370605},{"id":"https://openalex.org/C153180980","wikidata":"https://www.wikidata.org/wiki/Q19776675","display_name":"Commit","level":2,"score":0.5748999714851379},{"id":"https://openalex.org/C2777363581","wikidata":"https://www.wikidata.org/wiki/Q15098235","display_name":"Harm","level":2,"score":0.5577999949455261},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.535099983215332},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.47769999504089355},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.38339999318122864},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.3675999939441681},{"id":"https://openalex.org/C41065033","wikidata":"https://www.wikidata.org/wiki/Q2825412","display_name":"Adversary","level":2,"score":0.361299991607666},{"id":"https://openalex.org/C2778403875","wikidata":"https://www.wikidata.org/wiki/Q20312394","display_name":"Adversarial machine learning","level":3,"score":0.357699990272522},{"id":"https://openalex.org/C73301696","wikidata":"https://www.wikidata.org/wiki/Q5469984","display_name":"Formalism (music)","level":3,"score":0.3249000012874603},{"id":"https://openalex.org/C140547941","wikidata":"https://www.wikidata.org/wiki/Q7797194","display_name":"Threat model","level":2,"score":0.29339998960494995},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.28029999136924744},{"id":"https://openalex.org/C2988773926","wikidata":"https://www.wikidata.org/wiki/Q25104379","display_name":"Generative adversarial network","level":3,"score":0.27230000495910645}],"mesh":[],"locations_count":1,"locations":[{"id":"pmh:oai:dash.harvard.edu:1/42719210","doi":null,"is_oa":true,"landing_page_url":"https://dash.harvard.edu/handle/1/42719210","pdf_url":"https://dash.harvard.edu/bitstreams/5f162ba2-70b5-4276-9f86-f2c4fde46308/download","source":{"id":"https://openalex.org/S4306401540","display_name":"Digital Access to Scholarship at Harvard (DASH) (Harvard University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I136199984","host_organization_name":"Harvard University","host_organization_lineage":["https://openalex.org/I136199984"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Thesis or Dissertation"}],"best_oa_location":{"id":"pmh:oai:dash.harvard.edu:1/42719210","doi":null,"is_oa":true,"landing_page_url":"https://dash.harvard.edu/handle/1/42719210","pdf_url":"https://dash.harvard.edu/bitstreams/5f162ba2-70b5-4276-9f86-f2c4fde46308/download","source":{"id":"https://openalex.org/S4306401540","display_name":"Digital Access to Scholarship at Harvard (DASH) (Harvard University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I136199984","host_organization_name":"Harvard University","host_organization_lineage":["https://openalex.org/I136199984"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Thesis or Dissertation"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7115393376.pdf","grobid_xml":"https://content.openalex.org/works/W7115393376.grobid-xml"},"referenced_works_count":43,"referenced_works":["https://openalex.org/W1673923490","https://openalex.org/W1945616565","https://openalex.org/W2799194071","https://openalex.org/W2970290563","https://openalex.org/W3093235747","https://openalex.org/W3098267758","https://openalex.org/W4308023630","https://openalex.org/W4309395891","https://openalex.org/W4318718936","https://openalex.org/W4321177655","https://openalex.org/W4366330503","https://openalex.org/W4376312115","https://openalex.org/W4377121468","https://openalex.org/W4380353722","https://openalex.org/W4381713708","https://openalex.org/W4384648205","https://openalex.org/W4384918448","https://openalex.org/W4385019259","https://openalex.org/W4385374425","https://openalex.org/W4385387622","https://openalex.org/W4385436553","https://openalex.org/W4386436506","https://openalex.org/W4386437528","https://openalex.org/W4386721684","https://openalex.org/W4387355345","https://openalex.org/W4387390310","https://openalex.org/W4387436590","https://openalex.org/W4387839052","https://openalex.org/W4388067733","https://openalex.org/W4388329043","https://openalex.org/W4388555306","https://openalex.org/W4390190425","https://openalex.org/W4391334942","https://openalex.org/W4391766701","https://openalex.org/W4391800306","https://openalex.org/W4393157467","https://openalex.org/W4400315206","https://openalex.org/W4401042550","https://openalex.org/W4405181600","https://openalex.org/W4407661798","https://openalex.org/W7133204634","https://openalex.org/W7133214447","https://openalex.org/W7133254726"],"related_works":["https://openalex.org/W4399911261","https://openalex.org/W6960257819","https://openalex.org/W7153406348","https://openalex.org/W4414337173","https://openalex.org/W7162679401","https://openalex.org/W7215058211","https://openalex.org/W4399986452","https://openalex.org/W7081668605","https://openalex.org/W4391237900","https://openalex.org/W4386437528"],"abstract_inverted_index":{"The":[0,119],"advent":[1],"of":[2,122,150,181,199,214,230,270,328],"new":[3],"generative":[4,226,231],"AI":[5,17,64,81,117,132,160,185,200,248,345],"models,":[6,221,227],"such":[7,32,50],"as":[8,47,335,337],"OpenAI\u2019s":[9],"GPT-4,":[10],"has":[11,22,243],"seen":[12],"a":[13,24,110,147,179,326],"huge":[14],"increase":[15,26],"in":[16,27,30,72,141,164,196,289,313,357,374],"capabilities.":[18],"With":[19],"this,":[20,321],"there":[21],"been":[23],"corresponding":[25,339],"the":[28,75,87,142,174,197,205,241,246,252,287,295],"ways":[29,73],"which":[31,43,290],"models":[33,54,65,69,133,161,346],"can":[34,126,154,292],"be":[35,128],"used":[36],"to":[37,46,52,63,70,92,115,162,194,208,219,225,245,265,279],"commit":[38],"harm":[39,127],"by":[40],"malicious":[41,152],"actors,":[42],"we":[44,172,236,285,322],"refer":[45],"misuse.":[48,118,201],"One":[49],"method":[51],"misuse":[53,100],"is":[55,204],"through":[56,134],"adversarial":[57,97,108,135,190,215,217,238,315,330],"attacks.":[58],"Informally,":[59],"these":[60,280,348],"are":[61,104],"inputs":[62,156],"that":[66,74,145,157,261,307],"cause":[67,158],"said":[68],"act":[71],"developers":[76],"did":[77],"not":[78],"intend.":[79],"Most":[80],"systems":[82,103],"require":[83],"users":[84,153,291],"to,":[85],"at":[86],"very":[88],"least,":[89],"have":[90],"access":[91,244,294],"their":[93],"input.":[94],"Thus,":[95],"if":[96],"attacks":[98,109,218,224,239,278,331],"for":[99,184,258],"exist,":[101],"most":[102],"vulnerable.":[105],"This":[106],"makes":[107],"natural":[111],"and":[112,166,189,228,298,317,324,332,350,369,401],"important":[113],"setting":[114,288],"study":[116,198],"central":[120],"question":[121],"this":[123,140,203],"thesis":[124],"is:":[125],"committed":[129],"with":[130,363,380],"current":[131,159,344],"attacks?":[136],"We":[137,250,273,300,342],"decisively":[138],"answer":[139],"affirmative,":[143],"showing":[144],"across":[146],"wide":[148],"range":[149],"settings,":[151],"craft":[155],"misbehave":[163],"alarming":[165],"harmful":[167,271],"ways.":[168],"In":[169,233,282],"Chapter":[170,234,283,358,375],"2,":[171],"introduce":[173],"novel":[175,253],"Interaction":[176],"Context":[177],"Framework,":[178],"collection":[180],"formal":[182],"definitions":[183],"systems,":[186],"model":[187],"behaviors,":[188],"attacks,":[191],"specifically":[192],"designed":[193],"aid":[195],"Notably,":[202],"first":[206],"formalism":[207],"unify":[209],"three":[210],"previously":[211],"disparate":[212],"notions":[213],"attacks:":[216],"discriminative":[220],"prompt":[222],"injection":[223],"jailbreaking":[229],"models.":[232],"3,":[235],"explore":[237],"when":[240],"user":[242],"underlying":[247],"model.":[249],"present":[251],"Image":[254],"Behavior":[255],"Matching":[256],"Algorithm":[257],"creating":[259],"images":[260],"force":[262],"multimodal":[263],"LLMs":[264],"display":[266],"four":[267],"different":[268],"types":[269],"behaviors.":[272,281],"also":[274],"extend":[275],"existing":[276],"text-based":[277],"4,":[284],"consider":[286],"only":[293],"AI\u2019s":[296],"input":[297],"output.":[299],"create":[301],"Tensor":[302],"Trust,":[303],"an":[304,314],"online":[305],"game":[306],"pits":[308],"players":[309],"against":[310],"one":[311],"another":[312],"attack":[316],"defense":[318],"contest.":[319],"Using":[320],"collect":[323],"release":[325],"dataset":[327],"128,808":[329],"46,457":[333],"defenses,":[334],"well":[336],"two":[338],"robustness":[340],"benchmarks.":[341],"evaluate":[343],"on":[347],"benchmarks":[349],"find":[351],"all":[352],"perform":[353],"poorly.":[354],"Work":[355,372],"done":[356,373],"3":[359],"was":[360,377],"conducted":[361,378],"jointly":[362,379],"Euan":[364],"Ong,":[365,392],"Professor":[366],"Stuart":[367,402],"Russell,":[368],"Scott":[370],"Emmons.":[371],"4":[376],"Sam":[381],"Toyer,":[382],"Olivia":[383],"Watkins,":[384],"Ethan":[385],"Mendes,":[386],"Justin":[387],"Svegliato,":[388],"Tiffany":[389],"Wang,":[390],"Isaac":[391],"Karim":[393],"Elmaaroufi,":[394],"Pieter":[395],"Abbeel,":[396],"Trevor":[397],"Darrell,":[398],"Alan":[399],"Ritter,":[400],"Russell.":[403]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2025-12-15T00:00:00"}