Skip to content
LinkedInX

Security Frameworks

About 10 minutes

Prerequisites: Key Attack Techniques

To safely design and operate generative AI systems, organizations need to systematically evaluate risks using industry-standard frameworks. This page explains the major frameworks related to generative AI security.[1][3][5][6][7]

Major frameworks related to generative AI security can be categorized by purpose and target as follows.[1][3][5][6][7]

FrameworkPublisherMain PurposeTarget
OWASP LLM Top 10OWASPIdentifying major LLM application risksDevelopers, security engineers
NIST AI 600-1NIST (US)Generative AI risk managementOrganizations, developers
MITRE ATLASMITREKnowledge base of AI attack tacticsSecurity researchers, red teams
ISO/IEC 42001ISO/IECAI management system standardEntire organizations
NIST Privacy FrameworkNIST (US)Privacy risk managementEntire organizations

The OWASP LLM Top 10 (OWASP Large Language Model Top 10) is a guideline summarizing the major security risks of LLM applications. OWASP (Open Web Application Security Project) published the first version in 2023 and released the 2025 version in November 2024.[1][2]

RankRiskOverview
LLM01Prompt InjectionTaking control of the system through malicious prompts
LLM02Insecure Output HandlingVulnerabilities from using LLM output without validation
LLM03Training Data PoisoningManipulating model behavior by altering training data
LLM04Model Denial of ServiceDisabling model function through mass requests
LLM05Supply Chain VulnerabilitiesDependency risks from third-party models and libraries
LLM06Sensitive Information DisclosureUnintended exposure of training data or system prompts
LLM07Insecure Plugin DesignAttacks via plugins and tools
LLM08Excessive AgencyRisks from granting too much authority to agents
LLM09OverrelianceRisks from uncritical trust in LLM output
LLM10Model TheftUnauthorized acquisition of model internals or training data

This table summarizes OWASP’s 2023/2024 project page.[2]

The 2025 version updates areas related to agentic risks, external dependencies, RAG/vector databases, and output reliability.[1]

  • LLM06 (Excessive Agency): Guidance on permission granting for agents becomes more detailed
  • LLM07 (System Prompt Leakage): Changed from insecure plugin design to system prompt leakage
  • LLM08 (Vector and Embedding Weaknesses): Updated to cover weaknesses in vector databases and embeddings used by RAG systems
  • LLM09 (Misinformation): Organized around misleading or unreliable generative AI outputs

NIST AI 600-1 is a document published by NIST (National Institute of Standards and Technology) in July 2024 that extends the NIST AI Risk Management Framework (AI RMF) specifically for generative AI. It identifies 12 risks that are unique to or exacerbated by generative AI and organizes the priority concerns organizations should manage.[3]

Risk AreaOverview
CBRN Information or CapabilitiesEasier access to or synthesis of dangerous chemical, biological, radiological, or nuclear information or capabilities
ConfabulationConfidently stated but erroneous or false content, also called hallucination or fabrication
Dangerous, Violent, or Hateful ContentGeneration or spread of violent, illegal, self-harm, or hateful content
Data PrivacyLeakage, inference, disclosure, or de-anonymization of personal or sensitive data
Environmental ImpactsCompute, energy, and environmental impacts from training and operating GAI systems
Harmful Bias or HomogenizationAmplified social bias, performance disparities, or undesired output homogeneity
Human-AI ConfigurationAnthropomorphism, overreliance, automation bias, or emotional entanglement with AI systems
Information IntegrityGeneration or spread of content that fails to distinguish fact, opinion, fiction, or uncertainty
Information SecurityOffensive cyber enablement or compromise of training data, code, model weights, or systems
Intellectual PropertyUnauthorized production or replication of copyrighted, trademarked, licensed, or trade-secret material
Obscene, Degrading, and/or Abusive ContentHarmful abusive content, including CSAM and nonconsensual intimate imagery
Value Chain and Component IntegrationReduced transparency or accountability across upstream components, data, models, and suppliers

NIST AI 600-1 organizes suggested actions for generative AI risks along the four functions of NIST AI RMF (GOVERN/MAP/MEASURE/MANAGE).[3][4]

FunctionRole
GOVERNEstablish policies, processes, and accountability structures for AI risk management
MAPIdentify and categorize AI system context and risks
MEASUREAssess identified risks quantitatively and qualitatively
MANAGEImplement measures to mitigate, accept, transfer, or avoid risks

MITRE ATLAS (Adversarial Threat Landscape for AI Systems) is a knowledge base that systematically organizes adversarial attack tactics, techniques, and procedures (TTPs: Tactics, Techniques, and Procedures) against AI systems. MITRE develops and maintains it.[5]

ItemMITRE ATT&CKMITRE ATLAS
TargetGeneral cyberattacksAttacks on AI systems
FocusNetwork and endpointsML/AI models and pipelines
Main attackersAPT groups, malware authorsAI researchers, malicious actors
  • Reconnaissance: Gathering information about AI system configuration and models in use
  • ML Model Access: Establishing API access and black-box/white-box access
  • ML Attack Staging: Creating adversarial examples and preparing backdoor data
  • Impact: Triggering model malfunction, service disruption, and extracting confidential information

ISO/IEC 42001 is an international standard for the responsible development, operation, and management of AI systems. It defines management system requirements for organizations to manage AI systems.[6]

ItemISO 27001ISO/IEC 42001
TargetInformation security overallAI system management
FocusConfidentiality, integrity, availabilityResponsible use of AI systems
Main contentISMS requirementsAI management system (AIMS) requirements
Applicable organizationsAll organizations handling IT and informationOrganizations developing or using AI

The NIST Privacy Framework is a voluntary tool for organizations to manage privacy risks. In the context of generative AI, the following are especially important.[7]

  • Identify-P: Identify privacy risks by determining what personal data the generative AI handles
  • Govern-P: Establish privacy policies and organizational accountability structures
  • Control-P: Implement user controls over data collection, processing, and sharing
  • Communicate-P: Explain privacy practices to users with transparency
  • Protect-P: Implement data protection measures to mitigate privacy risks

Privacy Challenges Unique to Generative AI

Section titled “Privacy Challenges Unique to Generative AI”
  • Personal information reconstruction from training data: Models may be able to reproduce personal information (names, addresses, phone numbers) present in the text they were trained on.[8]
  • Membership inference attacks: Attacks that attempt to infer whether specific data was included in the training data
  • Differential Privacy: One technical countermeasure to privacy risk. A technique that uses statistical noise to make the influence of individual data harder to infer.[9]

  • OWASP LLM Top 10 is the most practical guideline for risk assessment during design and development
  • NIST AI 600-1 is a comprehensive framework for building organizational-level risk management structures
  • MITRE ATLAS is a knowledge base specialized for red teaming and threat intelligence
  • ISO/IEC 42001 is a systematic standard for organizations seeking international AI management certification
  • In practice, combining multiple frameworks is effective

Q: Which framework should I start with?

A: For developers and security engineers, OWASP LLM Top 10 is the best starting point. Risks are summarized in 10 items and can be used as a checklist during design and testing phases. NIST AI 600-1 is appropriate for organizations looking to establish overall governance structures, and ISO/IEC 42001 is suitable when considering international certification.[1][3][6]

Q: How should I distinguish between OWASP LLM Top 10 and NIST AI 600-1?

A: OWASP LLM Top 10 focuses on “what to prevent” (identifying specific risks) and serves as guidance for technical countermeasures. NIST AI 600-1 focuses on “how to manage organizationally” (establishing risk management processes) and is useful for building governance structures. A practical division is to have development teams primarily reference OWASP, while management and governance departments primarily reference NIST.[1][3]

  1. OWASP, OWASP Top 10 for LLM Applications 2025, November 17, 2024
  2. OWASP, OWASP Top 10 for Large Language Model Applications
  3. NIST, Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile (NIST AI 600-1), July 2024
  4. NIST, AI Risk Management Framework
  5. MITRE, MITRE ATLAS
  6. ISO, ISO/IEC 42001 - Artificial intelligence management system
  7. NIST, Privacy Framework
  8. Nicholas Carlini et al., Extracting Training Data from Large Language Models, USENIX Security 2021
  9. Cynthia Dwork, Differential Privacy: A Survey of Results, 2008
Quiz