Cybernion
← Insights

ISO 42001

Is ISO 42001 Mandatory in Australia?

16 Oct 2026 · Updated 16 Oct 2026

No. ISO 42001 is not mandated by any Australian law. AI governance here runs on the Voluntary AI Safety Standard, published in September 2024 with ten guardrails, and the government has consulted on mandatory guardrails for high risk AI. ISO 42001 certification stays voluntary, but it is a practical route to those guardrails and is increasingly asked for in procurement.

“Do we have to get ISO 42001?” is the first question Australian AI teams ask, usually after a customer or a board raises it. The short answer is no, there is no law. The longer answer is that the ground is moving, and the organisations treating AI governance as optional are the ones that will scramble later. What follows sets out the legal position, the guidance that does apply, and why voluntary does not mean ignore it.

No. ISO/IEC 42001:2023 is a voluntary international standard for an AI management system (AIMS), and no Australian legislation requires an organisation to hold it. It is certifiable, meaning an accredited body can audit the management system and issue a certificate on a three year cycle, but that certification is a choice, not a statutory obligation.

That puts it in the same position as ISO 27001 for information security: widely recognised, often asked for commercially, but not mandated by law for most organisations. Demand for ISO 42001 in Australia today is commercial and procurement driven, not regulatory.

What is the Voluntary AI Safety Standard, and how does it relate to ISO 42001?

What Australia has, instead of a law, is guidance. The Department of Industry, Science and Resources published the Voluntary AI Safety Standard in September 2024. It sets out ten guardrails for organisations developing and deploying AI:

  • accountability and governance
  • risk management
  • data governance
  • testing
  • human oversight
  • transparency with end users
  • contestability
  • supply chain transparency
  • record keeping
  • stakeholder engagement

The guardrails are voluntary guidance, not enforceable rules. They align closely with ISO 42001, so an AI management system is a practical way to put the guardrails into operation and keep the evidence that an organisation is following them. In the AI governance work we run, we treat the ten guardrails as the intent and the AIMS as the structure that makes them real and auditable.

Could AI governance become mandatory in Australia?

The direction of travel points that way, but nothing is settled. The government has consulted on mandatory guardrails for AI used in high risk settings, which signals an intent to move some obligations from voluntary to required. What those obligations will cover, which organisations they will bind, and when they will commence are open questions, and the scope and timing are not settled.

The practical reading is this. An organisation that stands up an AI management system now is building the accountability, risk assessment, human oversight and record keeping that any mandatory regime is likely to expect. If the rules harden, that organisation adjusts rather than starts from zero. This is the same pattern Australian firms saw with privacy and critical infrastructure obligations, where the ones that prepared early absorbed the change quietly.

Why get ISO 42001 certified if it is voluntary?

Because voluntary standards still get asked for. The trigger is almost always external: a customer wants assurance that the AI in a product is governed, a government tender asks how AI risk is managed, or a board wants evidence rather than a promise. An accredited ISO 42001 certificate answers that in a way a self declaration cannot, because an independent body has audited the system.

ISO 42001 also asks for something broader than information security. Its AI system impact assessment looks at the consequences of an AI system for individuals, groups and society, including fairness, transparency and safety, not only risk to the organisation. That is the part regulators and customers care about most, and it is the part hardest to show without a system behind it. For where to begin, the ISO 42001 readiness checklist sets out what to have in place, and the complete ISO 42001 guide walks through the full standard. We run ISO 42001 readiness and advisory for Australian AI product companies.

For organisations selling into Europe, the position is different again: see ISO 42001 and the EU AI Act, where a binding law does apply.

Frequently asked questions

Is ISO 42001 a legal requirement in Australia?

No. No Australian law requires ISO 42001. It is a voluntary international standard an organisation chooses to certify against. AI governance in Australia currently runs on the Voluntary AI Safety Standard, which is guidance rather than legislation.

What is the Voluntary AI Safety Standard?

It is guidance published by the Department of Industry, Science and Resources in September 2024, setting out ten guardrails for safe and responsible AI, covering accountability, risk management, data governance, testing, human oversight, transparency, contestability, supply chain transparency, record keeping and stakeholder engagement. It is voluntary, and it aligns closely with ISO 42001.

Will AI governance become mandatory in Australia?

The government has consulted on mandatory guardrails for AI in high risk settings, so the direction is toward more obligation, but the scope and timing are not settled. An organisation that builds an AI management system now is positioned for whatever becomes mandatory, rather than starting from nothing later.

Why get ISO 42001 certified if it is voluntary?

Demand is commercial and procurement driven. Customers, tenders and boards increasingly want independent proof that an organisation governs AI responsibly, and an accredited certificate answers that in a way a self declaration cannot. It also maps onto the Voluntary AI Safety Standard guardrails.

Does ISO 42001 satisfy the Australian AI guardrails?

Not as a formal legal mapping, because the guardrails are voluntary guidance. But the ten guardrails align closely with an ISO 42001 AI management system, so building the AIMS is a practical route to demonstrating them. We treat the guardrails as the intent and the AIMS as the evidence.


Written by Gaurav Vikash, an ASD endorsed IRAP assessor and senior cyber security leader with 18 years of experience across Australia, the UK and Asia, including CISO and senior security leadership roles. He holds CISSP, CISA, CISM and CRISC and is an ISO 27001 Lead Auditor and ISO 42001 Lead Implementer, and speaks regularly at industry conferences.

Talk to us.

Sources:

  1. ISO/IEC 42001:2023, Artificial intelligence management system, iso.org, 2023
  2. Voluntary AI Safety Standard, Department of Industry, Science and Resources, September 2024
  3. Safe and responsible AI in Australia, Department of Industry, Science and Resources, accessed October 2026

Last updated: 16 October, 2026

Related serviceISO 42001 ReadinessSee the service →

Talk to us about your engagement

A scoped proposal within one business day.