An Australian song hit No 1 on radio. Now people are asking if AI made it.

Key points

• Josh Fawaz's cover of Madonna's 'Like a Prayer' reached No 1 on Australia's National Radio Airplay chart in 2024.
• Music experts and working musicians publicly question whether the track was made using generative AI, software that produces audio or text from a simple typed prompt.
• Fawaz went from a little-known artist to a viral name within months, driven partly by thousands of radio spins.
• No verified independent analysis has yet confirmed or ruled out AI involvement in the track's production.

Josh Fawaz was, until recently, an obscure Australian producer. Then his cover of Madonna's 1989 classic 'Like a Prayer' started getting played. A lot. It reached the top spot on the National Radio Airplay chart, the tally of how many times a song gets broadcast across Australian stations each week.

Now the song is at the centre of a dispute that goes well beyond one artist's career.

Music industry figures and fellow musicians have started asking publicly whether a human actually made this. The concern, reported by The Guardian, is that the track may have been produced using generative AI, software that can compose and mix music without a person playing a single note. You feed it a prompt, and it builds you a song.

If that claim holds up, it matters on two fronts.

Chart positions shape which artists get label deals and touring opportunities. A track that quietly skips the human creative process but still collects the same rewards raises serious fairness questions for musicians who spent years developing their craft.

There's also a disclosure problem. No standard rule in Australia currently requires artists or labels to declare when AI tools made a meaningful contribution to a recording. Listeners and chart compilers had no way of knowing, if the allegations are correct.

None of that means Fawaz has done anything illegal. Generative AI tools sit in a legal grey area in most music markets. Using them isn't automatically fraud or copyright infringement, though that's being argued in courts across multiple countries as labels sue AI music companies.

What experts are calling for is transparency: a simple, consistent label, the way food packaging lists ingredients.

What does this mean for people who listen to the radio?

For now, it changes very little day-to-day. But it suggests that airplay counts, metrics the industry has always treated as signals of genuine human popularity, may need new rules to stay meaningful.

AI2Day has been covering AI transparency questions across sixteen stories in the past ninety days, and this case crystallises why disclosure rules matter: without them, the chart itself becomes an unreliable signal. The Australian music industry hasn't announced any formal review of how AI-assisted content is handled on the airplay chart. That conversation, though, is now very hard to avoid.

My read: the song may be perfectly legal, but a chart that can't tell listeners what they're hearing has already lost something it won't easily get back.