← All articles
RAG Security
2 articles
Agent SecurityAgentic PermissionsAI Agent SecurityAI Governance EvidenceAi ImpactAI Incident ResponseAi IntegrationAI Red TeamingAI SDLC & Product SecurityAI SecurityAI Security Engineer CareerAi Security EngineeringAI Security FoundationsAI Security MonitoringAI Security ToolsAI Supply ChainAI System InventoryArchitecture and Trust BoundariesAts SystemsAttackCareer DevelopmentCorporate CultureCorporate Culture And LeadershipCulture SecurityCyber SecuritycybersecurityCybersecurity StrategyData Exposure and PrivacyDefendDetection EngineeringDistributed GovernanceDistributed SystemsEconomic GovernanceEducationEvaluation and Regression TestingEvidenceEvidence Based GovernanceFuture of WorkgovernanceGovernance And ResilienceGovernance Evidence and Customer TrustGovernance, Risk & ComplianceHiring & TalentHiring StrategyIncident ResponseIncident Response & ObservabilityLeadership And GovernanceLLM Application SecurityLogging and TelemetryMapMLOps & Platform SecurityModel and Provider RiskModel Supply ChainOperational RiskOrganizational GovernanceOrganizational ResiliencePlatform GovernancePrivacy & Data ProtectionPrompt InjectionPrompt Injection & Context SecurityPsychological SafetypsychometricsRAG AuthorizationRAG SecurityRecruitment And TalentRed Teaming & Evaluationsred-teamseceng-workbenchSecure Architecture & DesignSecure RAGSecurity ArchitectureStochastic GovernanceStochastic ResilienceSystemic ResilienceTalent AcquisitionTalent EngineeringTeam EngineeringTechnical IntelligenceThreat ModelingToolchain IntegrityTraining & WorkshopsVendor Risk & ProcurementWorkforce ScienceWorkplace Evolution
Attack
RAG Data Leakage: How Private Documents Escape Through Retrieval, Embeddings, and Context Windows
RAG data leakage happens when retrieval, embeddings, metadata, prompt context, generated answers, logs, or deletion workflows expose information outside intended boundaries. Secure RAG requires authorization-aware retrieval, tenant isolation, metadata filtering, sensitive-data minimization, protected traces, retention limits, and incident-ready evidence.
12 min read
Attack
OWASP LLM Top 10 2025 Explained for Engineers Building Real AI Products
Teams adopt LLM features quickly and then discover that traditional AppSec checks miss retrieval abuse, tool misuse, and unsafe output handling. The Top 10 helps because it names the failure modes that need design and test work.
4 min read