Who regulates your use of AI?
Australia has no AI Act. Instead it has 62 obligations spread across 31 different bodies, most of which reach AI through duties that already existed. Pick your industry and see the ones that apply to you, what each actually requires, and where to read it yourself.
Choose your industry
Every row read off the issuing body's own website on 7 September 2026. Direct means the instrument names AI or automated decision-making. Indirect means it binds your use of AI through general duties that make no mention of it. This is general information, not legal advice. It tells you which door to knock on, not what to do behind it.
Of 62 instruments across 31 bodies, 38 name AI or automated decision-making explicitly and 24 do not mention it at all. The National AI Plan of December 2025 confirmed Australia will rely on existing technology-neutral law rather than a single AI statute, and the proposed mandatory guardrails for high-risk AI were dropped.
The absence of an AI Act is not the relief it sounds like. One statute would give you one compliance project. Instead every regulator that already governs you now reads its existing rules as covering your AI, which means the obligation arrives without a commencement date, a transition period, or a notice telling you it now applies to something new.
What your state actually adds.
Two things surprise most people. No state or territory has an Act regulating AI use by private business. And every state and territory privacy Act binds the public sector only, whatever you have been told. The state layer that really reaches you is surveillance and health records law, and it is wildly uneven.
Verified 7 September 2026 against each jurisdiction's own legislation register or the responsible department's own page. Rows marked as binding the public sector only are shown so you can rule them out, because businesses are routinely told they apply. General information, not legal advice.