Artificial Intelligence

AI Threat Modelling and Risk Management Services

Anchoram’s AI Threat Modelling and Risk Management Services help organisations identify, assess, prioritise and mitigate cybersecurity threats and operational risks associated with the development, deployment, integration and use of Artificial Intelligence (AI) systems.

Anchoram provides structured, risk-based threat modelling and security risk management services to help organisations understand their AI attack surfaces, identify potential threat actors and attack pathways, assess security control effectiveness and establish appropriate safeguards throughout the AI system lifecycle.

How we deliver

Our Approach

Our services encompass AI applications, underlying models, training and inference environments, data pipelines, cloud infrastructure, third-party AI services, Retrieval-Augmented Generation (RAG) architectures, APIs and agentic AI workflows. We combine established cybersecurity threat modelling techniques with AI-specific adversarial threat analysis, architecture reviews, security testing and risk assessment methodologies.

Our approach is informed by internationally recognised standards, frameworks and industry guidance, including:

  • NIST AI Risk Management Framework (AI RMF 1.0) and the NIST Generative AI Profile (AI 600-1), supporting structured identification, measurement, management and governance of AI-related risks.
  • NIST Cybersecurity Framework (CSF 2.0) and NIST guidance on adversarial machine learning, supporting cybersecurity risk management and assessment of AI-specific attack techniques.
  • Australian Signals Directorate (ASD) / Australian Cyber Security Centre (ACSC) guidance, including the Guidelines for Secure AI System Development, Engaging with Artificial Intelligence and the Essential Eight, where applicable.
  • MITRE ATLAS (Adversarial Threat Landscape for Artificial-Intelligence Systems), supporting identification and analysis of adversarial tactics, techniques and attack scenarios targeting AI-enabled systems.
  • OWASP Top 10 for Large Language Model Applications and associated AI security guidance, supporting identification and mitigation of vulnerabilities in generative AI applications and AI integrations.
  • ISO/IEC 27001, ISO/IEC 27005, ISO/IEC 23894 and ISO/IEC 42001, supporting information security management, security risk assessment, AI risk management and AI management system governance.
  • Australian Government Information Security Manual (ISM) and the Protective Security Policy Framework (PSPF), where relevant to government information, systems and operating environments.

Outputs and Outcomes

The following key deliverables, tailored to client requirements:

  1. AI Threat Modelling and Attack Surface Assessment: Documented AI system architectures, critical assets, data flows, trust boundaries, threat actors, attack scenarios, potential vulnerabilities and attack pathways, mapped against recognised frameworks such as MITRE ATLAS and OWASP.
  2. AI Security Risk Assessment and Risk Register: Comprehensive assessment of identified AI security threats, including likelihood, impact, existing control effectiveness, inherent and residual risk ratings, risk ownership and prioritised treatment recommendations.
  3. AI Security Controls and Mitigation Framework: Recommended security architectures, technical safeguards, access controls, model and data protection measures, AI-specific security testing requirements, monitoring mechanisms and incident response controls aligned with recognised standards and leading practices.
  4. AI Threat Validation and Security Testing Report: Results of agreed adversarial testing, AI red teaming, vulnerability assessments, prompt injection testing, model security assessments and control validation activities, including identified weaknesses, supporting evidence and remediation recommendations.
  5. AI Risk Treatment Roadmap and Continuous Monitoring Framework: Prioritised remediation initiatives, implementation recommendations, risk management procedures, monitoring metrics, reporting dashboards and continuous improvement mechanisms to support ongoing AI security risk management.

Anchoram’s services help organisations embed security-by-design principles into AI initiatives, strengthen their resilience against emerging AI threats and establish practical, repeatable processes for managing AI security risks across the technology lifecycle.

Partners and Membership

DAMA Australia
Certifer
Robinson Ryan
ISACA
The Institute of Internal Auditors Australia

Obtaining further
Information by making Contact
with our Experienced staff.

We’re available for 8 hours a day!
Contact to require a detailed analysis and assessments of your plan.