The Price of Inference: A Longitudinal Economic Analysis of Hierarchical LLM Credential Leakage

The Price of Inference: A Longitudinal Economic Analysis of Hierarchical LLM Credential Leakage | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The Price of Inference: A Longitudinal Economic Analysis of Hierarchical LLM Credential Leakage Tanishq S, Vignesh B This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8802620/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The integration of Large Language Models (LLMs) into the software supply chain has fundamentally altered the nature of credential leakage. Unlike static secrets (e.g., database passwords), LLM API keys are liquid economic assets—direct bearers of computational reasoning and token quotas. This paper introduces the Theory of Economic Credential Stratification, identifying a critical divergence in how these assets are managed compared to their utility. Using CHRONOS, a custom-built forensic instrument, we analyzed exposed artifacts across GitHub. We report a paradoxical “Protection Gap”: Tier 1 (GPT-4) credentials—despite having high abuse potential for LLMjacking [5]—exhibit a mean survival time (¯tsurv) of 48 hours in nonproduction artifacts, significantly longer than lower-value keys. Furthermore, we identify “Dataset Poisoning” as a systemic blind spot: valid credentials embedded in .jsonl and .parquet training files often persist indefinitely, becoming part of the model’s latent knowledge. Artificial Intelligence and Machine Learning Information Retrieval and Management Software Engineering Systems and Networking Analysis Operations Research Development Economics Other Economics LLM Security Credential Leakage Supply Chain Security Forensics Economic Stratification Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8802620","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":586798244,"identity":"dc0a28f1-f0a7-4a34-87db-68f9d8d8e59e","order_by":0,"name":"Tanishq S","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8ElEQVRIiWNgGAWjYLCCBIYDDGzyjw8AmRIyhJWzQbXwM6QlgLTwEKeFAahFsiHHAMQirEV+fvOzBw/+3JHdcODM51c3aix4GNgPH92AT4vBMTZzg8S2Z8YbDvZus845BnQYT1raDbxa2BjMJBIbDiduOMy7zTiHDahFgscMrxb5NvZvEgl/gFqO8TwzzvlHhBaGYzxmEglshxNn9vAwP85tI0KLwbGcMonEtsPG/RJsZsy5fRI8bIT8It98fJvkjz+HZdskmB9/zvlWJ8fPfvgYfodBAWMDMIokQCw2YpTDtDB/IFb1KBgFo2AUjCwAAAH4S1lkL7l7AAAAAElFTkSuQmCC","orcid":"https://orcid.org/0009-0009-3344-1658","institution":"","correspondingAuthor":true,"prefix":"","firstName":"Tanishq","middleName":"","lastName":"S","suffix":""},{"id":586798245,"identity":"74ada0a9-7aab-4bea-84ad-ea0b3463b6cd","order_by":1,"name":"Vignesh B","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Vignesh","middleName":"","lastName":"B","suffix":""}],"badges":[],"createdAt":"2026-02-06 04:46:38","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-8802620/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8802620/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102296719,"identity":"b2b181ed-b412-4320-833f-4e676fc55e2e","added_by":"auto","created_at":"2026-02-10 10:21:04","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":188202,"visible":true,"origin":"","legend":"","description":"","filename":"Paper.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8802620/v1_covered_c6118c81-1a15-4734-9da0-216f3bbce8b3.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eThe Price of Inference: A Longitudinal Economic Analysis of Hierarchical LLM Credential Leakage\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"LLM Security, Credential Leakage, Supply Chain Security, Forensics, Economic Stratification","lastPublishedDoi":"10.21203/rs.3.rs-8802620/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8802620/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe integration of Large Language Models (LLMs) into the software supply chain has fundamentally altered the nature of credential leakage. Unlike static secrets (e.g., database passwords), LLM API keys are liquid economic assets—direct bearers of computational reasoning and token quotas. This paper introduces the Theory of Economic Credential Stratification, identifying a critical divergence in how these assets are managed compared to their utility. Using CHRONOS, a custom-built forensic instrument, we analyzed exposed artifacts across GitHub. We report a paradoxical “Protection Gap”: Tier 1 (GPT-4) credentials—despite having high abuse potential for LLMjacking [5]—exhibit a mean survival time (¯tsurv) of 48 hours in nonproduction artifacts, significantly longer than lower-value keys. Furthermore, we identify “Dataset Poisoning” as a systemic blind spot: valid credentials embedded in .jsonl and .parquet training files often persist indefinitely, becoming part of the model’s latent knowledge.\u003c/p\u003e","manuscriptTitle":"The Price of Inference: A Longitudinal Economic Analysis of Hierarchical LLM Credential Leakage","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-09 08:16:25","doi":"10.21203/rs.3.rs-8802620/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f8396e50-dfd4-4e0e-970e-e68ea99a1e25","owner":[],"postedDate":"February 9th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":62435006,"name":"Artificial Intelligence and Machine Learning"},{"id":62435007,"name":"Information Retrieval and Management"},{"id":62435008,"name":"Software Engineering"},{"id":62435009,"name":"Systems and Networking"},{"id":62435010,"name":"Analysis"},{"id":62435011,"name":"Operations Research"},{"id":62435012,"name":"Development Economics"},{"id":62435013,"name":"Other Economics"}],"tags":[],"updatedAt":"2026-02-09T08:16:25+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-09 08:16:25","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8802620","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8802620","identity":"rs-8802620","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}