Quick hit

On 5 August 2025, the Productivity Commission (PC) released its interim report, Harnessing data and digital technology.

The PC has called for submissions on the use of copyright materials to train AI models, whether copyright reform is required, and a proposal to introduce a fair dealing exception for text and data mining.

Submissions are due by 15 September 2025.

Background to the interim report

In 2023, the Australian Government set out its “five pillar” productivity growth agenda. In late 2024, the PC was asked to complete five inquiriesto identify reforms under each of the five pillars.

One of the Government’s growth pillars is “Harnessing data and digital technology”.In that context, the PC is examining the economic potential of AI and the impact of AI on productivity.

The issues have also been the subject of separate consideration, including:

  1. In December 2023, the Government established a Copyright and Artificial Intelligence Reference Group to engage with stakeholders on issues at the intersection of AI and copyright.
  2. In November 2024, the Senate Select Committee on Adopting Artificial Intelligence recommended that the Government undertake consultation with the creative industry to consider an appropriate mechanism to ensure fair remuneration is paid to creators for the use of copyright material to train AI systems.

Key copyright issues in relation to generative AI

The copyright issues in relation to generative AI arise at two stages: input and output.

At the input stage, the key copyright question is whether the unlicensed training of a Generative AI model involves an infringement of copyright. This involves first, the technical question of whether the training process in fact involved any reproduction of the copyright works, and second, the legal question of whether any such reproduction falls under an exception to infringement.

Exceptions to copyright infringement are different in different countries. Unlike Australia, the United States has a “fair use” defence. Many generative AI companies are seeking to rely upon “fair use” for unlicensed training of AI models. The availability of a fair use defence is currently being hotly contested in a number of proceedings brought by copyright owners in the US.

At the output stage, key questions are:

  • whether an output infringes copyright, and if so, whether the provider of the Generative AI model is responsible for that infringement; and
  • whether the output of generative AI is protected by copyright, which involves questions of whether there is sufficient human authorship.

The PC’s focus has been on the input question.

What does the Productivity Commission say about training AI models?

The PC noted the need to reproduce copyright materials (at least temporarily) to train AI models, and that there is evidence to suggest that large AI models are already being trained on copyright materials without consent or compensation.

The PC referred to concerns that the Australian copyright regime is not keeping pace with the rise of AI technology, “whether because it does not adequately facilitate the use of copyrighted works or because AI developers can too easily sidestep existing licensing and enforcement mechanisms.”

The PC set out three policy options:

  1. no change, with copyright owners able to enforce existing rights using existing mechanisms;
  2. measures to facilitate the licensing of copyright materials (such collective licensing); and
  3. introducing a fair dealing exception for text and data mining.

What does the PC say about text and data mining?

In its interim report, the PC noted the absence of a TDM exception from Australian law and seeks feedback on the likely effects of a TDM exception on the AI market, the creative sector and productivity in general. The PC noted that the TDM exception is not a “blank cheque”, because the use must be considered “fair”.

The PC referred to the availability of TDM exceptions in several other jurisdictions, each of which is different. The PC referred to the current consideration of TDMs in the UK (part of the UK Government’s consultation on copyright and AI).

What is the PC asking for submissions on?

The PC has specifically sought feedback as follows:

The PC is seeking feedback on the issue of copyrighted materials being used to train AI models.

Are reforms to the copyright regime (including licensing arrangements) required? If so, what are they and why?

The PC is also seeking feedback on the proposal to amend the Copyright Act 1968 (Cth) to include a fair dealing exception for text and data mining.

  1. How would an exception covering text and data mining affect the development and use of AI in Australia? What are the costs, benefits and risks of a text and data mining exception likely to be?
  2. How should the exception be implemented in the Copyright Act – for example, should it be through a broad text and data mining exception or one that covers non-commercial uses only?
  3. Is there a need for legislative criteria or regulatory guidance to help provide clarity about what types of uses are fair?

Next steps

Submissions on the interim report are due by 15 September 2025.

The Productivity Commission will consider those submissions, and its final report is due to be provided to the Government by 13 December 2025.

For assistance with copyright infringement and copyright reform issues, please contact Adam Simpson, Clare Young, Ian McDonald or Anna Spies.