Emerging technology regulation
A conceptual introduction to how existing Australian copyright and privacy frameworks are being tested by data-driven and AI-related technologies, presented as developing principle rather than settled doctrine.
Learning outcomes
- Identify examples of how data-driven and AI-related technologies are testing existing copyright and privacy frameworks in Australia.
- Explain why this area should be approached as evolving principle and policy debate rather than settled, exhaustive doctrine.
- Distinguish applying settled intellectual property doctrine from reasoning carefully about genuinely unresolved technology-law questions.
The other articles in this module describe long-settled doctrine. This one does not. It looks at how courts, regulators and Parliament are currently working out how existing copyright and privacy principles apply to new fact patterns produced by data-driven and AI-related technologies. Some of what appears here may be answered differently, or answered for the first time, within the life of a law degree — and recognising that honestly is itself part of the skill this article teaches.
Why this topic is different from the rest of the module
Copyright, patents, trade marks and confidential information are each built on doctrine that has been tested and refined over a long period. Emerging technology regulation is not a sixth body of settled doctrine sitting alongside them; it is an emerging application of those same frameworks to circumstances they were not originally designed for. A student should be able to tell the difference between citing settled authority for a proposition and describing an area of law that is still being worked out — and should say so explicitly, rather than presenting an unsettled position with false confidence.
Copyright and data-driven technologies
Data-driven and AI-related technologies raise a cluster of copyright questions that existing doctrine does not obviously answer. Does using copyright material to build or train a data-driven system involve a reproduction of the kind the Copyright Act is concerned with — one of the exclusive rights in s 311 — and if so, does any existing exception cover it? Can output generated by such a system be an original work at all, given Australian copyright law's traditional assumption that a work has a human author? If so, who owns it — the person who built the system, the person who operated it, or no one? Were the existing fair dealing exceptions, framed around purposes like research, review and news reporting, designed with this kind of large-scale automated use in mind at all? None of these questions has a single settled answer at present; they are being worked through by courts, regulators and law reform processes, and a careful student states them as open questions rather than asserting a conclusion the law does not yet support.
Privacy and data-driven technology
A closely related set of questions arises under Australia's federal privacy framework, which sits alongside intellectual property law rather than within it. Large-scale collection and processing of personal information to build or operate data-driven systems raises questions about consent, purpose limitation and accountability that are conceptually distinct from copyright's concern with protecting expression — but the two frequently intersect in practice, for example where the same dataset raises both a copyright question about the material used and a privacy question about the personal information it contains.
Confidential information and data
The law of confidential information is also being tested by these technologies. Where a system is trained on a proprietary dataset, questions arise about whether information fed into the system, or able to be extracted back out of it, retains the necessary quality of confidence, and whether using confidential information without authorisation to build or operate such a system could itself found an action for breach of confidence.
How a student should approach this area
The right approach is not to memorise current positions as though they were fixed, but to build a method: identify which settled doctrinal framework — copyright, confidential information, privacy — a new technology fact pattern is being tested against; identify precisely which element of that settled framework is genuinely contestable in the new context; distinguish a real legal question from a policy question that is properly addressed to Parliament or a regulator rather than a court; and remain alert to genuine developments in legislation and case law rather than treating any position described here as final.
Applying this in a problem question
- Identify which established doctrinal framework — copyright, confidential information, privacy, or more than one — the new-technology fact pattern engages.
- State the settled elements of that doctrine first, using the relevant sibling article as your reference point.
- Identify precisely which element is contested or unclear when applied to the new technology, rather than treating the whole area as uncertain.
- Distinguish a genuinely open legal question from a policy question that is really addressed to Parliament or a regulator.
- Signal the uncertainty honestly in your answer, rather than asserting a settled conclusion the current law does not support.