For most of music history, making a finished record usually meant developing musical skills first. You learned an instrument, trained your voice, worked with producers or built enough technical knowledge to turn an idea into sound.

Generative music introduces another route. Someone can now arrive with years of experience in film, writing, design or another creative field, then discover that many of those skills transfer surprisingly well into music. What if some of the people who become unusually good at making AI music do not currently think of themselves as musicians at all?

Adam Hartwiński is an interesting example. He is a Polish filmmaker whose professional background is in directing rather than traditional music education, yet he is now using Suno v6 to build HARTVINSKY, a dark electronic project whose 17-track debut album TWO WORDS releases worldwide on 16 October 2026.

TWO WORDS leans toward dark electronic and deep-house production, with compact, cinematic arrangements and hook-driven structures. What makes the project interesting is not simply that a filmmaker is making music. It is how directly the instincts Hartwiński developed in film have carried into the way he directs AI-generated sound.

A new route into music

Hartwiński grew up between Poland and Sweden, studied film in Sweden and later graduated in directing from the Warsaw Film School. His previous work includes FIRST… / PO PIERWSZE…, which he says won Best Foreign Short at the Arizona International Film Festival, and THE BALLERINA / BALETNICA, which won the Grand OFF award for Best Polish Film. His professional work in Polish film and television also includes JONNA. UCIECZKA DO RAJU.

That background affects how he approaches music. Rather than beginning with conventional production questions, Hartwiński often thinks about pacing, tension, anticipation and payoff.

“I think of a track as a scene. The opening sound is the first shot. A repeated motif is a visual motif. A silence before a drop is a cut. A return of the hook is a callback.”

A filmmaker already spends years deciding what an audience should notice, when information should arrive and when something has gone on too long. Generative music gives those instincts somewhere new to go.

The same could apply elsewhere. A writer may understand dialogue and hooks, a visual artist may have a strong sense of identity, and a game designer may understand atmosphere and repetition. None of those skills automatically make someone good at music, but AI makes it easier to test whether they translate.

Directing instead of starting from a blank prompt

Hartwiński rarely begins by asking Suno to simply create a good song. The full track may not exist in his head, but one important element usually does.

It might be a vocal rhythm, a whistle, a short hook or a particular emotional situation. He might decide that the hook needs to arrive immediately, that the vocal should feel close rather than theatrical, or that the arrangement should briefly empty out before something returns.

The point is to narrow the problem rather than hand the model the identity of the song. Suno can then explore possibilities around a smaller creative target.

That distinction becomes more important as music generation improves. If models can produce convincing songs across thousands of styles, the harder question becomes less about whether they can make something impressive and more about what should be made.

This connects closely with previous interviews. Vitalii Klimenko argued in AI Music and the New Burden of Choice that generation creates options while production creates decisions. Brian Myk similarly described in What Skills Should Artists Value in the Age of AI Music? how easier experimentation can increase the importance of judgment and selection.

Hartwiński reaches a similar conclusion from film. His experience choosing between possibilities did not begin in a recording studio. It came from directing and editing.

The creative skill may increasingly be rejection

Hartwiński describes Suno as a “generative instrument”, but his process is built around iteration rather than accepting the first output. He defines a target, generates, listens, rejects, adjusts and compares versions until something fits.

“The key creative act is often rejection.”

That line points to a broader change in AI production. When creating another version becomes cheap, the ability to say no becomes more important.

A generation can be technically impressive and still be wrong. It may be too bright, too busy, too sentimental or simply too generic for the project. Editors and directors are already trained to remove good material when it weakens the whole, which may explain why those skills transfer so naturally into generative workflows.

Musical talent may become broader than musicianship

There is an important distinction between performing music and making good musical decisions. Someone can have a strong instinct for when a hook should arrive, how long a section should last or whether a vocal feels convincing without being able to perform those elements themselves.

Historically, that gap could stop an idea from becoming a finished record. Generative tools reduce part of that barrier.

This does not make instrumental skill or vocal performance less meaningful. It means another kind of ability becomes easier to see. Someone who was never technically capable of producing the music they imagined may discover that they are very good at directing it.

That could broaden the pool of people entering music. The next interesting AI creator might come from film, advertising, theatre, gaming or visual art rather than a band, conservatory or bedroom-production background.

Unlimited possibility can make identity harder

The freedom offered by generative music also creates a problem. One creator can move rapidly between house, rock, folk, orchestral music and experimental electronic production.

“Unlimited choice is actually the enemy of identity.”

Hartwiński says TWO WORDS only became coherent because he limited what HARTVINSKY was allowed to be. The album stays dark rather than bright, favors compact structures and repeated motifs, and avoids much of the vocabulary he associates with festival EDM.

The consistency also goes beyond sound. Every track is built around a two-word phrase that works as a dramatic trigger, such as a command, provocation, confession or decision whose meaning can change as the song develops.

In WATCH ME, a confident demand for attention eventually becomes a plea directed at one person who matters. In ONE MORE, an ordinary phrase becomes the language of addiction. YOU WISH turns defensive flirtation into something more vulnerable when “you wish” becomes “I wish.”

The album therefore has consistency at several levels. Its production belongs to the same world, its titles share a language, and many of the songs follow a similar dramatic structure built around setup, tension and reversal.

That mirrors another theme from our interviews. In The Story Behind AI Music Artists You Might Have Missed, Doniell Clay discussed maintaining identity across artificial performers. Hartwiński is solving a similar problem through sound and narrative.

Film does not only shape the sound

Hartwiński's film background affects more than arrangement. It also shapes what happens inside the songs.

One principle he brings directly from filmmaking is “late entry, early exit”. A scene should begin as late as possible and end before it becomes redundant. That helps explain why HARTVINSKY tracks tend to be compact.

His approach to suspense works in a similar way. Hartwiński refers to Alfred Hitchcock's famous example of the bomb under the table, where tension comes from the audience knowing something important is waiting to happen.

Music can create the same expectation. Establish a motif, remove it and delay its return, and the listener begins waiting for it. The eventual payoff works because the audience has already learned what to expect.

The same idea appears in the album's two-word phrases. A phrase can return later with a different meaning because the listener already understands what it meant before.

These techniques are not new to music. What is notable is how easily someone trained in another medium can now apply them directly to music production.

AI may blur the borders between creative careers

That could have wider consequences for the music industry. Generative tools may make movement between creative fields much easier.

A filmmaker can make an electronic album. A songwriter can develop finished recordings without becoming a performer. A visual artist can build an audiovisual project where music becomes another part of the same creative world.

This means AI could increase the supply of music in two ways. Existing musicians can make more, while people who previously lacked the technical route into music can begin making it at all.

That creates more competition, but also a larger talent pool. Someone who previously needed musicians, studio time and production knowledge to test an idea may soon be able to find out quickly whether their creative instincts work in music.

Some of those people will discover that they are not particularly good at it. Others may uncover abilities they had no practical way to express before.

This is why the AI music discussion may need to move beyond asking whether someone is a “real musician”. A more useful question is what creative skill the person is contributing.

A pattern is forming across AI music

Lunar Boom's recent interviews increasingly point toward the same conclusion from different directions.

Clay has focused on identity and storytelling. Klimenko has emphasized judgment when generation creates too many options. Myk has described how easier experimentation can make selection more important. Other creators have shown how AI can help move an idea toward a finished record without requiring the person behind it to perform every part.

Hartwiński adds another possibility. Some of the skills that become valuable in AI music may have been developed outside music entirely.

His background gave him experience with pacing, editing, suspense and narrative structure before HARTVINSKY existed. Generative music gave those skills another medium.

AI may not only change what existing musicians can do. It may change who discovers they are capable of making music in the first place.

HARTVINSKY becomes the test

TWO WORDS releases worldwide on 16 October 2026. Its 17 tracks are not built around a single AI experiment, but as a complete concept album following the same sonic and narrative rules throughout.

Hartwiński describes the transition simply.

“I don’t experience HARTVINSKY as leaving film to become a musician. I experience it as directing with a different material.”

That may be the broader lesson. As generative tools lower the technical barriers to finished music, some of the most interesting new creators may come from careers that previously had little to do with making records.

AI is changing what machines can do, but it may also change who gets to discover that they are good at making music.

For readers curious to hear the project, STILL HERE is a good starting point. Listen to HARTVINSKY — STILL HERE or watch the video preview.

TWO WORDS contains 17 tracks and releases worldwide on 16 October 2026.