The conversation about AI and songwriting has been dominated by extremes — either AI is going to replace songwriters entirely, or it’s a creative abomination that should be banned. The reality, as usual, is somewhere in between and more interesting than either extreme suggests.

I’ve talked to a dozen Australian songwriters and producers about how they’re actually using AI tools in their creative process. The answers surprised me.

What People Are Actually Doing

Overcoming Writer’s Block

The most common use case, mentioned by almost every songwriter I spoke to, is using AI as a brainstorming partner.

“I’ll feed it a theme or a first line and ask for twenty different directions I could take it,” a Melbourne-based songwriter told me. “I never use the AI’s output directly. But seeing those twenty options shakes something loose in my own brain. It’s like having a brainstorming session with a collaborator who’s not very good but very fast.”

This approach treats the AI as a creative prompt generator rather than a writing tool. The songwriter’s own sensibility and voice still drive the final product.

Melody Exploration

Several producers described using AI tools that generate melodic ideas based on chord progressions. They’ll input a harmonic structure and let the AI suggest melodic contours, then use those suggestions as starting points for their own melodies.

“It’s like having a really basic sketch that I can then turn into something real,” one producer explained. “The AI melodies are never good enough on their own — they’re too generic, too predictable. But they give me a shape to react to.”

Lyric Editing

A few songwriters use AI to help with the craft aspects of lyric writing — identifying awkward phrasing, suggesting rhyme schemes, checking metre. This is similar to using a spell checker for prose writing.

“I wrote the lyrics completely myself, then used AI to highlight lines where the syllable count was off or the imagery was weak,” a Sydney songwriter said. “It’s an editing tool, not a writing tool. The difference matters to me.”

Arrangement Suggestions

Producers are using AI tools that analyse a track and suggest arrangement ideas — where to add instruments, where to strip things back, how to structure a song for maximum impact. Again, these suggestions are starting points rather than final decisions.

What People Aren’t Doing

None of the songwriters I spoke to use AI to generate complete songs. None of them pass off AI-generated lyrics as their own. None of them rely on AI for the creative core of their work.

This might be self-selection — songwriters who are comfortable talking to a music journalist about AI use are probably the ones using it responsibly. But the pattern is consistent: AI as assistant, not author.

The Ethical Questions

Even when AI is used as a tool rather than a replacement, ethical questions arise.

Credit and Disclosure

Should artists disclose when AI tools were part of their creative process? There’s no industry standard here yet. APRA AMCOS hasn’t issued formal guidance on AI’s role in songwriting (though they’re reportedly working on it). The current position is essentially: if a human songwriter created the work with AI assistance, the human gets the credit and the royalties.

This seems reasonable for the current use cases — brainstorming, editing, arrangement suggestions. But as AI tools become more capable, the line between “assistance” and “co-creation” will get harder to draw.

Awards and Recognition

The ARIAs and other awards haven’t addressed AI use in submissions. Should a song that was partially developed using AI tools compete in the same categories as one that wasn’t? Most songwriters I spoke to think yes, as long as the AI’s contribution was at the tool level rather than the creative core. But opinions on where that line sits vary.

The Economics

If AI tools allow one songwriter to be as productive as three, what happens to the other two? This is the deeper economic question that the music industry hasn’t grappled with yet.

In the short term, AI tools are increasing individual productivity without eliminating jobs. Songwriters who use AI brainstorming produce more starting ideas, which leads to more finished songs, which potentially crowds the market further. Whether this is good or bad depends on your perspective.

Where I Stand

I think AI songwriting tools are legitimate creative aids when used honestly. A songwriter who uses AI to generate brainstorming options and then creates something original from that process has done the creative work. The tool facilitated the process but didn’t replace it.

I draw the line at presenting AI-generated content as original human creation. If an AI wrote the melody or the lyrics with minimal human input, that should be disclosed. Not because AI-assisted music is inherently bad, but because audiences and industry assessors deserve to know what they’re evaluating.

The music industry needs to develop clear frameworks for AI use in creative processes. APRA AMCOS, the ARIA, and artist organisations should be leading this conversation. Waiting for the technology to force their hand is a mistake.

For now, the Australian songwriters I know are using AI thoughtfully and honestly. That gives me some confidence that the creative community will navigate this transition better than the panic merchants suggest.

But we need the frameworks in place before the tools become powerful enough to make the ethical questions truly difficult.