Skip to the main content.
Contact Insicon Cyber
Linkedin YouTube

AI Security and Governance

Australian and New Zealand organisations are deploying AI faster than their governance frameworks can keep pace. Insicon Cyber AI Security and Governance is a practice of three connected services that address AI risk from every angle: finding and fixing vulnerabilities in AI systems, building the governance framework regulators require, and maintaining compliance across all obligations on an ongoing basis.

Secure AI. Governed AI. Compliant AI.
Five Eyes

AI Risk Is Now a Board Issue. What APRA, ASIC and the Five Eyes Are Telling Leaders in Australia and New Zealand.

In just eight weeks, three authoritative bodies issued AI risk directives aimed squarely at boards and executives across Australia and New Zealand.

On 22 June 2026, the heads of six cyber security agencies across the Five Eyes nations, including Australia's ASD and New Zealand's NCSC, told business leaders the window between AI-accelerated vulnerability discovery and exploitation is shrinking, and the timeline for action is months, not years.

Watch the replay: what APRA, ASIC and the Five Eyes expect on AI risk

Recorded on 14 July 2026. Matt Miller, Co-Founder, CEO and Fractional CISO, and Greg Bunt, Co-Founder, Director and Fractional CISO, explain what APRA, ASIC and the Five Eyes cyber security agencies are asking of boards and leadership teams in Australia and New Zealand, how the EU AI Act reaches local organisations, and where ISO 42001 fits. Practical, and focused on what to do next.

What is AI Security and Governance?

AI Security and Governance is the discipline of protecting, certifying, and continuously managing the security and compliance posture of artificial intelligence systems. It covers three distinct but connected disciplines: adversarial security testing of AI models and agents, implementation of an internationally certified AI Management System, and ongoing managed compliance across cybersecurity and AI governance frameworks.

For organisations in Australia and New Zealand operating in regulated sectors - education, financial services, aged care, and healthcare - AI Security and Governance is how boards demonstrate that AI adoption is controlled, auditable, and defensible.

Three services. One practice. One team.


AI Assurance

Test it. Protect it.

Expert-led adversarial testing of your AI systems - models, applications, and agents - powered by F5 AI Red Team. Real-world attack scenarios including prompt injection, jailbreak chaining, data exfiltration, and multi-agent privilege escalation. Findings feed directly into F5 AI Guardrails for runtime protection. Led by Insicon Cyber's experienced team. Executed by the Insicon Cyber aSOC.

Available as a one-off assessment (AI Assurance: Assess) or an ongoing subscription with continuous testing and always-on runtime protection (AI Assurance: Continuous).

Learn about AI Assurance

ISO 42001 - AI Management System

Govern it.

Implementation of ISO/IEC 42001:2023, the international standard for AI Management Systems. Insicon Cyber guides Australian and New Zealand organisations through gap assessment, AIMS development, policy and process design, and certification readiness. Built on the same rigour as our ISO 27001 practice. Mapped to APRA CPS 234, CPS 230, and the Australian Privacy Act 1988.

Available as a gap assessment, a full implementation engagement, or combined with Managed Compliance for post-certification maintenance.

Learn about ISO 42001

Managed Compliance

Maintain it. Permanently.

Ongoing management of cybersecurity and AI compliance obligations - Essential Eight, ISO 27001, ISO 42001, and NZISM - under one team, one reporting rhythm, one monthly investment. Continuous evidence collection, regulatory change monitoring, audit preparation, and board-ready compliance reporting. Insicon Cyber's fractional CISOs can participate in board risk committees and audit committees as your compliance accountability layer.

Learn about Managed Compliance

How organisations use AI Security & Governance

The three services work together but can be engaged independently. The most common starting points:

Starting with AI security

AI systems already in production. No AI-specific security testing in place. Recommended path: AI Assurance: Assess to establish the attack surface baseline, then AI Assurance: Continuous for ongoing protection, then ISO 42001 to build the governance framework around the findings.

Starting with governance

Board or regulator has requested AI governance evidence. ISO 27001 already certified. Recommended path: ISO 42001 Gap Assessment, then ISO 42001 Implementation, transitioning to Managed Compliance post-certification, with AI Assurance providing the technical risk evidence that feeds the AIMS.

Expanding an existing compliance programme

Already on Insicon Cyber Managed Compliance for Essential Eight or ISO 27001. AI systems being added to the environment. Recommended path: Managed Compliance scope expansion to include ISO 42001, combined with an AI Assurance: Assess engagement to baseline AI security posture.

Why Insicon Cyber for AI Security & Governance

Fractional CISO leadership on every engagement

Insicon Cyber AI Security & Governance is led by experienced cybersecurity, governance, and risk professionals. Every service engagement, board briefing, and audit committee presentation is led by a CISO-level practitioner, not a junior consultant.

SOC-executed, technology-led

The Insicon Cyber aSOC operates 24/7 across Australia and New Zealand. AI Assurance extends that operational capability into AI-specific adversarial testing and threat detection, using F5 AI Red Team and F5 AI Guardrails. Australian data sovereignty maintained throughout.

ANZ regulatory expertise built in

Every service is calibrated to the specific obligations of Australian and New Zealand regulated sectors: APRA CPS 234 and CPS 230, the Australian Privacy Act 1988, the NZ Privacy Act 2020, ASD Essential Eight, NZISM, and the ISO 42001 standard. Global frameworks interpreted for the ANZ market — not adapted from a Northern Hemisphere template.

ISO 27001 certified organisation

Insicon Cyber is ISO 27001 certified. The governance rigour we apply to our own operations is the same standard we implement for clients. When we guide your organisation through ISO 42001 certification, we do so as practitioners, not theorists.

AI Security & Governance and ANZ regulatory obligations

Regulatory expectations for AI governance are developing rapidly in Australia and New Zealand. Insicon Cyber monitors these obligations and maps service delivery to them.

APRA CPS 234 - Information Security

CPS 234 requires APRA-regulated entities to maintain information security capabilities commensurate with their risk profile, including the risk posed by material service providers. AI systems embedded in regulated environments - whether built internally or provided by third parties - are subject to CPS 234 assessment. AI Assurance provides the technical security evidence CPS 234 assessments require.

APRA CPS 230 - Operational Resilience

CPS 230 requires APRA-regulated entities to identify, manage, and test operational resilience, including the resilience of AI systems under stress. AI Assurance operational stress testing - covering latency, resource exhaustion, and availability under adversarial pressure - provides the documented evidence CPS 230 resilience assessments require.

Privacy Act 1988 (Australia) and NZ Privacy Act 2020

Data exfiltration through prompt injection is a live privacy risk for any organisation processing personal information through AI systems. AI Assurance adversarial testing includes data exfiltration scenarios specifically designed for Privacy Act-exposed environments. ISO 42001 implementation includes data governance controls and AI system impact assessments aligned to Privacy Act accountability obligations.

ISO/IEC 42001:2023 - AI Management System

ISO 42001 is the international standard for responsible AI management. Certification demonstrates to boards, regulators, and clients that an organisation has implemented a structured framework for AI risk assessment, system impact assessment, and continual improvement. Insicon Cyber implements ISO 42001 that build on proven ISO 27001 methodologies.

Start with a conversation

Talk to one of Insicon Cyber's fractional CISOs about your AI security and governance requirements. No-cost initial briefing. Scoped to your regulatory environment and sector.

Contact Insicon Cyber

Frequently asked questions

 

What is AI Security and Governance?

AI Security and Governance is the practice of protecting AI systems from adversarial attacks, governing their use through internationally recognised standards such as ISO 42001, and maintaining compliance with AI and cybersecurity frameworks on an ongoing basis. Insicon Cyber offers AI Security and Governance as a three-service practice for organisations in Australia and New Zealand.

Which organisations in Australia and New Zealand need AI Security and Governance?

Any organisation deploying AI in a regulated environment should have a structured AI Security and Governance programme. This includes APRA-regulated entities in financial services subject to CPS 234 and CPS 230, aged care and healthcare providers subject to the Australian Privacy Act 1988 or the New Zealand Privacy Act 2020, and organisations that are ISO 27001 certified or pursuing ISO 42001 certification.

How is AI security different from traditional cybersecurity?

Traditional cybersecurity controls, such as firewalls, web application firewalls and conventional penetration testing, were designed for deterministic systems with predictable inputs. AI systems are non-deterministic. Attacks such as prompt injection, jailbreak chaining and multi-agent privilege escalation are semantic problems that traditional controls can't detect. AI security needs a different class of testing and a different class of runtime protection.

What is prompt injection and why does it matter for organisations in Australia and New Zealand?

Prompt injection is an attack technique where a malicious input manipulates an AI system into behaving in an unauthorised way, such as disclosing sensitive data, bypassing access controls or executing unintended actions. It is one of the most common and most dangerous AI-specific attack vectors. For organisations processing personal information through AI systems, a successful prompt injection attack can trigger breach notification obligations under the Australian Privacy Act 1988 or the New Zealand Privacy Act 2020, and regulatory scrutiny from APRA for regulated entities in Australia.

What is ISO 42001 and is it mandatory in Australia or New Zealand?

ISO/IEC 42001:2023 is the international standard for AI Management Systems (AIMS). It establishes a structured framework for responsible AI development, deployment and governance. ISO 42001 is not currently mandatory in Australia or New Zealand, but it is increasingly expected by major clients, insurers and procurement frameworks. For APRA-regulated entities, ISO 42001 certification provides demonstrable evidence of AI governance capability aligned to CPS 234 and CPS 230 obligations.

How long does ISO 42001 implementation take?

ISO 42001 implementation typically takes 12 to 20 weeks from gap assessment to certification readiness, depending on the number of AI systems in scope and the maturity of existing governance frameworks. Organisations already certified to ISO 27001 typically reach readiness faster, because foundational ISMS structures can be extended rather than rebuilt. Insicon Cyber delivers ISO 42001 implementation in four phases: gap assessment, AIMS development, certification readiness and post-certification maintenance. Certification itself is issued by an accredited certification body.

What compliance frameworks does Insicon Cyber Managed Compliance cover?

Insicon Cyber Managed Compliance covers the Essential Eight, ISO 27001, ISO 42001 and NZISM. Clients can engage Managed Compliance for a single framework or across multiple frameworks under a unified programme. Every programme includes continuous evidence management, regulatory change monitoring, audit preparation and quarterly board-level compliance reporting.

What is F5 AI Red Team and how does Insicon Cyber use it?

F5 AI Red Team is an automated adversarial testing platform that simulates real-world AI attacks, including prompt injection, jailbreak chaining, data exfiltration and multi-agent privilege escalation, at scale and at speed. Insicon Cyber uses F5 AI Red Team as the core testing.