OWASP Top 10 for LLM Applications 2026
LLM02:2026 Sensitive Information Disclosure
23 advisories on ShieldMap fall under it through CWE-359, CWE-212, CWE-532, and 3 technique pages explain how the attacks work.
OWASP Top 10 for Agentic Applications
- ●ASI06:2026 Memory & Context PoisoningPersistent agent memory corruption leading to cross-session data leakage
CWE weaknesses
- ●CWE-200 Exposure of Sensitive Information to an Unauthorized ActorThe entry defines disclosure as exposing confidential, regulated, or privileged data through an unauthorized channel.
- ●CWE-359 Exposure of Private Personal InformationPII and PHI disclosure sits at the center of the entry, anchored to GDPR and HIPAA obligations across its examples.
- ●CWE-212 Improper Removal of Sensitive Information Before Storage or TransferRedaction and removal failures surface sensitive spans, such as text hidden under a black-rectangle redaction layer resurfaced by summarization.
- ●CWE-532 Insertion of Sensitive Information into Log FileObservability pipelines log full prompts, completions, and reasoning traces, exposing sensitive data through log and telemetry stores.
- ○CWE-285 Improper AuthorizationRetrieval runs before any authorization check, so cosine similarity returns documents the requester has no right to see.
- ○CWE-732 Incorrect Permission Assignment for Critical ResourceUnscoped drives, legacy permissions, and over-permissive knowledge bases feed the retrieval corpus with sensitive data.
- ○CWE-201 Insertion of Sensitive Information Into Sent DataTool-call arguments and outbound requests to external providers carry more sensitive fields than the task actually requires.
MITRE ATLAS tactics
- ●AML.TA0010 ExfiltrationMembership inference, model inversion, model extraction, and API-based data theft all move protected data out of the system, sometimes through covert DNS or image channels.
- ○AML.TA0013 Credential AccessAn embedded vendor API key placed in the system prompt can be coaxed out of the model, exposing a live credential.
- ○AML.TA0000 AI Model AccessExtraction, membership inference, and inversion run offline at unbounded rates against open-weights deployments, so the access mode itself is the attack surface.
- ○AML.TA0007 Defense EvasionCross-lingual, Base64, and hex encodings defeat regex and blocklist data-loss filters, and cross-modal transformation slips data past single-modality inspection.
NIST AI 600-1 risks
- ●NIST Data PrivacyThe entry centers on privacy leakage and de-anonymization of PII, PHI, biometric, and genomic data, including membership inference that confirms a named individual and defeats erasure obligations.
- ●NIST Information SecurityIt concerns confidentiality of the system and its data through model and data exfiltration, credential exposure, and extraction attacks.
- ○NIST Intellectual PropertyTrade secrets and model weights are protected assets at risk, and models can reproduce copyrighted or watermarked training material verbatim.
- ○NIST Value Chain and Component IntegrationThird-party observability platforms, SDKs, and integrated browser components disclose sensitive data through the surrounding ecosystem.
NIST AI RMF categories
- ○MAP 5 characterize impacts on peopleDisclosure is framed as impact to data subjects, where membership inference yields a regulatory-breach determination about a named individual under GDPR or HIPAA.
- ○MEASURE 2 evaluate trustworthinessA release gate requires disclosure red-teaming that quantitatively measures extraction, membership inference, embedding inversion, internal-state inversion, side channels, and LoRA extractability.
- ○MANAGE 4 risk treatment, response & recoveryThe disclosure incident-response playbook sets breach-notification timelines and treats leaks with unlearning, retraining, withdrawal, and vector and cache cleanup, plus vendor notice.
OWASP GenAI Data Security
- ●DSGAI01 Sensitive Data LeakageThe direct peer: unauthorized exposure of confidential, regulated, or proprietary data through any output channel is the whole risk.
- ●DSGAI18 Inference & Data ReconstructionMembership inference, embedding inversion, and internal-state inversion reconstruct protected data from measurable model behavior without receiving the content directly.
- ○DSGAI15 Over-Broad Context Windows & Prompt Over-SharingUnscoped drives, legacy permissions, and auto-appended full records feed the model more sensitive data than the task needs, the upstream over-sharing that drives most incidents.
- ○DSGAI08 Non-Compliance & Regulatory ViolationsThe disclosure carries the regulatory consequence, triggering EU AI Act, GDPR, HIPAA, and CCPA obligations and their breach-notification clocks.
- ○DSGAI14 Excessive Telemetry & Monitoring LeakageObservability platforms log full prompts, completions, retrieved chunks, and reasoning traces by default, exposing sensitive content to anyone with dashboard access.
CSA AI Controls Matrix domains
- ●DSP Data Security and Privacy Lifecycle Management
- ●IAM Identity & Access Management
- ○CEK Cryptography, Encryption & Key Management
- ○MDS Model Development Security
- ○AIS Application & Interface Security
Mappings and their reasons from the OWASP Top 10 for LLM Applications 2026, Appendix A (OWASP GenAI Security Project), shared under CC BY-SA 4.0. Mappings restructured as data; the ATT&CK section and the AICM relevance text left out. Navigation, not a compliance claim.