Australia Bans Largely AI-Generated Songs From Its Music Charts
Australia's music charts, run by the Australian Recording Industry Association (ARIA), will no longer include songs that are largely or entirely created by artificial intelligence. Under a new code effective from the Friday following the August 25 announcement, every release must be "substantially human made" to qualify for chart placement.
The rule change was triggered by a specific incident. An Australian DJ, Josh Fawaz, released an AI-assisted cover of Madonna's "Like a Prayer" that topped the ARIA dance singles chart and reached number two on Australia's overall singles chart. It racked up more than 48 million streams on Spotify alone and became a fixture on commercial radio. Fawaz added generative AI credits only after public backlash, by which point the track had already achieved major commercial success.
ARIA framed the decision as both a cultural and commercial safeguard. Chief executive Annabelle Herd said the changes reflect the organisation's aim to remain dynamic and "promote the human nature of artistry" in a fast-changing landscape. ARIA said Australian artists are competing for attention "in the most crowded market in history" and that it was "not interested in promoting or celebrating the success of AI-generated music that does not contain human artistry." Herd drew a sharper line on wholesale AI generation, arguing that music produced by services built on artists' recordings without licensing — meaning without permission or payment to the original creators — was a different matter entirely. A chart rewarding unlicensed AI output, she said, would undercut the very foundation of the recorded music ARIA exists to represent.
The new rules are not a blanket prohibition on AI tools. AI assistance remains permissible for tasks like mastering (the final polishing stage of a recording), drum machines, and auto-tune. The threshold for eligibility is that humans must have written the song and performed the lead vocal and primary instruments. Artists will be required to declare AI use when submitting a song for chart consideration, and ARIA said it could retrospectively adjust chart positions if it later emerged that music was largely AI-generated. The organisation may also request that awards given for reaching number one be returned. Artists and their representatives will have the right to challenge exclusion under the new framework. BBC News
Fawaz's AI-generated rework of "Like a Prayer" will be removed from ARIA chart eligibility under the new rules, according to the Australian Financial Review. The track's commercial success, achieved before its AI origins were disclosed, crystallised concerns within the Australian industry about transparency and the competitive displacement of human creators. Electronic act Peking Duk publicly backed ARIA's move against AI-generated music. AFR
The ARIA code is not an isolated national decision. It is based on a directive from the International Federation of the Phonographic Industry (IFPI), the global body representing the recorded music industry. In July, the IFPI said AI guidelines would be rolled out in Latin America, the Middle East, Africa, and Southeast Asia for use in their official charts. ARIA's updated code also excludes music that raises "stream or chart manipulation concerns," broadening the scope beyond AI provenance to address wider integrity issues in the streaming era. The precedent for chart exclusion on AI grounds was set earlier in 2025, when Sweden banned a song from its music charts because it was created by AI.
The Australian government's posture has been moving in the same direction. In July 2024, Prime Minister Anthony Albanese promised "the strongest possible protection" from AI for Australian creatives, pledging that writers and musicians would be able to choose whether and how their work was used in AI training. Albanese called the loss of that control and non-payment "theft." That commitment from the executive branch, combined with ARIA's chart-level enforcement, gives Australia one of the more coordinated national responses to AI-generated music among major recorded-music markets, though the legislative detail behind Albanese's pledge has yet to be fully tested.
The broader context here is about where the line falls between permissible assistance and impermissible generation. ARIA's framework, which keys eligibility to human authorship of the song and human performance of lead vocals and primary instruments, is workable but leaves grey zones. This is especially true for genres where production, sampling, and electronic sound design are inseparable from the composition itself. The retrospective adjustment provision and the demand for return of number-one awards introduce enforcement teeth that go beyond declarative policy. Whether chart compilers in other jurisdictions will adopt comparable enforcement mechanisms, or settle for lighter-touch disclosure requirements, will shape how much practical weight the IFPI directive carries.
The declaration requirement is the enforcement hinge. If artists must self-report AI use at the point of submission, the system depends on voluntary honesty, with retrospective audit as the backstop. That model places significant investigative burden on ARIA or its data partners to verify claims of human authorship, particularly for releases where AI tools have been used across multiple production stages in ways that are technically permitted but difficult to disentangle from the impermissible. The challenge clause for excluded artists provides due process, but the evidentiary standard for proving "substantially human made" creative output has not been publicly defined.
For now, ARIA has drawn a line that is clearer in principle than in application. The Fawaz case provided the catalyst; the IFPI directive provided the framework. What is unresolved is whether the human-authorship threshold can hold as generative AI tools become more deeply embedded in standard production workflows, and whether other national chart systems will follow Australia's enforcement-first approach or pursue softer regulatory paths.
