FOCUS Magazine · 15 Mar 2026
FOCUS came to the studio with a deceptively simple test: build a complete song – lyrics, melody, a singer, a finished video – and do it as fast as possible. The result said less about music than about how quickly an idea now turns into finished content.
1 in 3
new streaming tracks are now fully AI-generated
97%
of listeners can’t tell an AI song from a human one (Ipsos)
~200k
songs uploaded every day – roughly 140 a minute
The pipeline behind that test is the part worth understanding – not because the tools are magic, but because of where the effort actually goes:
- Brief the conceptA language model reads the brief, profiles the audience, and drafts the lyrics.
- Generate the musicA few minutes in a music model turns that lyric into a finished, mixed track.
- Write the storyline, cast the singerThe model proposes a narrative and an avatar to carry it.
- Render the videoA video model performs the song in short clips, stitched into a single cut.
- Review and refineBy the next day the rough cut is there – the slow part is now taste, not logistics.
The fascinating thing about AI is that you can realize any song and any screenplay with it. Budget plays only a secondary role.Björn Schneider, in FOCUS (translated)
What actually changes isn’t the toolset, it’s the economics. Whether a scene is set in London, Rio or Shanghai – or at the North Pole, or underwater – barely changes the budget any more, and effects that once meant animating frame by frame now don’t. Production costs fall sharply, especially for TV spots – not to zero, but far enough that budget becomes a secondary concern. Concept, storyboard, edit and post still run the classic way; what’s now fully AI-driven is the execution of scenes and animation. And when making something is no longer the hard part, the hard part moves upstream: to the idea, and to the first two seconds.
That is where the real fight is now. We’re in a battle for attention you can often only win with AI – if there’s no wow in the first two seconds, people are already gone. Speed isn’t the prize; it’s the price of entry.
Speed like that deserves a hard question, and I’d rather ask it than wave it away. A study by cognitive scientist Claudia Roswandowitz found the brain still tells the difference – synthetic voices light up its reward system a little less than real ones. I take that seriously.
But I read it differently than the sceptics do. It measures a gap, not a ceiling – and that gap is the narrowest it has ever been, closing with every model generation. So it isn’t an argument for waiting; it’s the argument for judgment. The studios that win won’t be the ones who held back, but the ones who put these tools in the hands of people who know what actually moves an audience. AI doesn’t replace that instinct – it hands it a far bigger lever.
Read the full FOCUS feature, “Fake-Bands, Fake-Musik” → focusplus.de ↗
