Australia’s music industry has officially had enough. Starting this week, the Australian Recording Industry Association, better known as ARIA, has updated its eligibility rules to exclude songs that are largely or wholly produced by artificial intelligence from appearing on its official charts. The new standard is straightforward: any release that wants a shot at the top spots must be, in ARIA’s own words, “substantially human made.”
What the New Rules Actually Say
The updated code is not a blanket ban on technology in the studio. Artists can still lean on AI tools for specific tasks like mastering tracks, and traditional production aids such as drum machines and auto-tune remain perfectly acceptable. The hard line, however, is this: a human must have written the song, performed the lead vocals, and played the primary instruments. When the creative core of a piece of music belongs to a machine rather than a person, that piece is no longer chart-eligible.
ARIA was pointed in its language when announcing the change, stating that Australian artists are competing for attention in what it called “the most crowded market in history.” The association made clear it has no interest in promoting or celebrating AI-generated music that strips out human artistry entirely. The message to the industry is unambiguous: if a person’s hands, voice, and creative soul are not at the centre of the work, it does not belong on the chart.
The Madonna Cover That Lit the Fuse
Rules rarely change in a vacuum, and this one is no different. The catalyst was a cover of Madonna’s iconic “Like A Prayer” by Australian DJ Josh Fawaz. The track climbed to the top of the ARIA dance singles chart and reached number two on the country’s overall chart, all while racking up more than 48 million streams on Spotify alone. It became a fixture on commercial radio playlists, the kind of ubiquitous track you hear in shopping centres and taxis without quite knowing why.
Then came the backlash. Questions surfaced about just how much of the song was the product of generative AI rather than human performance. Fawaz subsequently added generative AI credits to the release, a move that only intensified scrutiny rather than quieting it. That controversy, reported in detail by BBC News, became the tipping point that pushed ARIA to formalise what many in the industry had already been quietly demanding.
Why This Matters Beyond Australia
It would be easy to file this story under “local regulatory quirk” and move on. That would be a mistake. What Australia has done is something no other major music market has tried with this kind of institutional weight behind it: it has codified a definition of human artistry and attached real commercial consequences to it. Chart placement is currency in the music business. It shapes streaming algorithms, radio rotation, festival bookings, and media coverage. Locking AI-heavy music out of that ecosystem sends a financial signal that artists and labels everywhere will feel.
The broader context matters too. Generative AI music tools have become increasingly sophisticated and accessible over the past few years. Platforms exist that can produce a full song, complete with vocals, lyrics, and instrumentation, in minutes, for virtually no cost. That democratisation sounds appealing in theory, but it creates a structural problem: human musicians who spend years developing their craft are being asked to compete with software that has no overhead, no creative block, and no bad days.
The Question of Authenticity in a Digital Age
There is a philosophical dimension here that goes deeper than chart eligibility. Music has always been a human communication, a way of transmitting emotion, experience, and cultural identity from one person to another. When the origin of that communication is an algorithm trained on existing human work, something in that exchange changes. Whether that change is acceptable is a debate that artists, listeners, and policymakers are only beginning to have seriously.
ARIA’s position is that it wants to “promote the human nature of artistry,” and that phrasing is worth sitting with for a moment. It implies that artistry is, by definition, a human quality. Not every observer will agree. There are serious thinkers who argue that creativity is a process, not a property of biological beings, and that dismissing AI-generated work on principle is a form of category snobbery. That debate is legitimate and worth having.
But for now, Australia’s charts have a clear answer. And given that the Fawaz track hit number two nationally before anyone noticed its AI provenance, the timing of that answer feels overdue rather than premature.
What Comes Next for Artists and Labels
For artists who use AI purely as a production tool, a shortcut for mastering or a way to demo ideas quickly, very little changes. The rules as written are designed to protect creative originality without outlawing every piece of technology in a modern studio. That is a relatively considered approach, and it gives the industry room to adapt without throwing out legitimate innovation.
For labels, the pressure is now on due diligence. Signing a track and pushing it to radio without understanding its creative origins is a reputational risk that the Fawaz episode illustrated vividly. Expect contracts, release agreements, and licensing paperwork to start including AI disclosure clauses as standard practice.
And for listeners, the shift raises a question worth chewing on: when you streamed that Madonna cover 48 million times, did the experience feel any different knowing a machine may have been doing the singing? If your honest answer is no, then the conversation about AI and authenticity in music is far more complicated than any single rule change can resolve.
So here is the real question this moment asks of all of us: is a song’s value in how it was made, or in how it makes you feel, and do we actually have the courage to decide?
