A working professional’s view from inside the room

Hi everybody. Hope you’re doing marvellously well.

There is an overwhelming amount of noise right now about artificial intelligence and music. Predictions, panic, hot takes, and endless theorising from people who often are not actually making records for a living. So let’s cut through that and talk about what is genuinely happening, right now, from inside the industry.

I am incredibly fortunate to do music for a living. I am making records, mixing albums, speaking daily with producers, engineers, mixers, songwriters, mastering engineers, and artists. This is not speculation. This is observation.

The first real impact: generic music is already being replaced

The earliest and most obvious casualty of AI is generic music. That work is already disappearing.

A close friend of mine is a mastering engineer who does a huge amount of sync work. Film, television, documentaries, commercials. A large company, and I will not name them, recently let go of both full time staff and long term independent contractors who were producing soundalike instrumental tracks for TV and film.

They replaced them with prompts.

That is not a theory. That is happening now.

AI can already generate music that is good enough for background documentary cues, lifestyle television, filler content, and certain commercials. No royalties. No fees. No musicians to pay. From a business perspective, it is brutally simple.

If your work lives primarily in the world of disposable background music, AI is not coming. It has already arrived.

Songwriting is changing faster than most people realise

This is the part many people still underestimate.

Songwriters are using AI to generate ideas every single day. Full songs. Chord progressions. Melodies. Structures. Styles. Some people use them directly. Others manipulate the results heavily.

Even where platforms have limited stem exporting, there are workarounds. Tracks can be separated, analysed, and rebuilt. Real musicians then replay the parts. Real singers replace the vocals. Producers shape and refine the material.

You are already hearing songs that began life as AI generated ideas.

There will be more music released than ever before. Successful artists will release more material than they ever have, because the initial idea generation phase is faster. That is the reality.

This is not about one novelty AI song topping a chart. That was simply the headline moment. The quieter truth is far more significant.

This is happening now with major label artists

It is crucial to be absolutely clear about this.

This is not future facing. This is not experimental. This is happening now, including with major label artists.

Major artists are already using AI assisted workflows to accelerate output. That means more releases, more singles, more versions, and more content, at a pace that would have been unthinkable even a few years ago. Even when AI generated material is heavily rewritten, replayed, or refined, it often forms the foundation of the creative process.

The end listener usually has no idea where the song truly began.

Why this consolidates success at the top

The knock on effect is profound.

Artists who already have an audience will be able to release far more music, far more frequently. Algorithms reward consistency and volume, so this increased output further reinforces their visibility. Success becomes even more concentrated.

At the same time, it becomes harder for new artists to break through the noise. Not because they lack talent, but because the sheer volume of professionally marketed content is overwhelming.

The ladder is being pulled further up.

The dangerous psychological effect on new artists

This is the part that worries me the most.

New artists looking at this landscape may conclude that the only way to survive, let alone succeed, is to adopt the same AI driven approach. Faster. Louder. More often. More polished. More generic.

That belief is deeply dangerous.

When everyone uses the same tools to chase the same outcomes, uniqueness disappears. Music stops being expressive and becomes optimised. Creativity turns into a race rather than a conversation.

Why perfection culture made this inevitable

We need to be honest about something uncomfortable.

Our obsession with perfection is one of the reasons AI fits so easily into modern music production.

For years we have normalised perfectly tuned vocals, perfectly aligned drums, perfectly edited performances, and identical sounding records across genres. We have criticised singers for being slightly out of tune. We have mistaken character for incompetence.

When perfection becomes the baseline, machines thrive.

The records that stopped ageing well

There is a reason many records from the late 1990s and early 2000s have not aged gracefully. The early dominance of DAWs, grid editing, sample replaced drums, hyper compression, and rigid tuning created music that was technically impressive but emotionally brittle.

Today anyone can achieve a massive wall of guitars. Anyone can programme perfect drums. Anyone can lock performances to a grid. These are no longer creative differentiators.

What is difficult is capturing something human.

Why imperfection matters more than ever

I have always loved instruments with personality. I have a slightly out of tune piano in my Los Angeles studio that people love to criticise. They tell me I should replace it with a pristine virtual instrument.

I never will.

Each key sounds slightly different. It wobbles. It breathes. It has character. That character is irreplaceable.

This is why I have always championed artists who prioritised feel over flawlessness. From jazz musicians to punk icons, from technically astonishing composers to three chord truth tellers.

The part AI cannot replace

AI can generate acceptable music.

What it cannot do is tell the truth.

As John Lennon once said to my dear friend and mentor Jack Douglas, “Tell the truth and make it rhyme.”

That is still the job.

Disposable music will increasingly belong to machines. That is fine. Let it. Rise above it.

Write lyrics that mean something. Capture performances that feel dangerous, emotional, and alive. Play your instrument. Sing with conviction. Embrace flaws when they serve the song.

And remember this. You still need to perform. You still need to connect with human beings in a room. AI cannot walk on stage and move an audience.

So what should we actually do?

Use AI as a tool if it helps you. Absolutely. That is sensible.

However do not aim for perfection. Aim for individuality. Aim for truth. Aim for music that could only have been made by you.

AI will win at imitation.

It will never win at authenticity.

Do not let it win.