The conventional take on artificial intelligence and entertainment is that it threatens both movies and music about equally. Cheap content. Endless variations. Algorithms eat the artists. The take is half right. Movies are vulnerable in a way music is not, and the reason has nothing to do with which form of art is "more creative." It's about supply, distribution, and what people actually pay for.
The two industries look superficially similar. They both make recorded media. They both run on subscription platforms. They both have famous people whose names sell tickets. Below the surface, they have almost nothing in common. The differences decide what artificial intelligence can do to them.
Movies Have a Supply Bottleneck
The American film industry releases about one hundred and fifty wide films a year, which works out to about three a weekend. Add limited releases, foreign distribution, and streaming originals, and the catalog of new releases is somewhere in the low thousands per year. The amount of stuff competing for attention at the top of the funnel is small. The biggest movie of the year usually has a budget of two hundred million dollars and a marketing campaign approaching the same number. Most weekends in the year, you can name the top three movies playing without looking it up.
Now imagine a world where artificial intelligence can produce a watchable feature length movie for ten thousand dollars in compute. The implication isn't that audiences suddenly want to watch ten thousand machine made movies a year. The implication is that the bottleneck of "what's playing" gets blown apart. The reason theaters and streaming platforms can extract economics today is that the supply of movies is constrained by who can finance, cast, and produce them. If a credible film can be conjured out of a prompt, the constraint goes away. Not for the prestige theatrical release. For the long tail. The Saturday night you might have rented a thriller that turns out to be exactly your taste, written and rendered for an audience of one. The middle of the catalog gets flooded.
That's a real threat. The thing that makes a movie expensive is the same thing that makes the catalog feel curated. Take the cost out and the curation collapses. The big tentpoles probably survive because part of their value is shared experience. Everyone seeing Inside Out 2 the same weekend is part of why Inside Out 2 made six hundred and fifty million dollars. The middle, the procedural thrillers and rom coms and competently produced action movies, is where machine made content can substitute for the human made version, and where the consumer often won't notice or care.
Music Has the Opposite Problem
Spotify takes in something like one hundred thousand new tracks a day. The catalog is already so deep that no human will ever hear most of it. The supply of music is effectively infinite, and the supply of new music is effectively infinite, and adding more music to the pile, by humans or by machines, has very little effect on what people listen to.
That sounds like the same problem as movies, just bigger. It isn't. The reason is that the economics of music don't run on recordings anymore. They run on touring.
For most working musicians, recordings are a marketing channel. The album exists so the artist has something to play on stage. The streaming royalties are real but small. The tickets, the merchandise, and the meet-and-greets are where the money lives. The top tier of the music industry is a touring industry now. Taylor Swift's Eras Tour grossed more than two billion dollars. The recorded music side of her business, important as it is, is dwarfed by what happens in stadiums.
Artificial intelligence can flood the recorded side. It can fill background playlists. It can generate workout music indistinguishable from human made workout music. It cannot fill an arena. The thing the audience is paying for at a concert is not the recording. It's the room, the people, the moment, the artist they care about being thirty feet away. None of that is reproducible by software.
That sets a floor under human musicians that doesn't exist for human filmmakers. The cheap end of the catalog gets eaten. The expensive end is protected by physics, geography, and ritual. The artist who wants to make a living can. They have to tour, but they could already do that, and most of them already were.
What Movies Don't Have
Movies don't have an arena tour. The closest analog is the press junket and the premiere, neither of which are revenue events for the studio. The relationship between the audience and the movie is through the screen. If the screen can be filled by software, the relationship can be filled by software too.
You could imagine an industry response. Live filmed events. Theatrical experiences with stunts and Q and A's and the cast in attendance. The thing is, that already exists. It's called Broadway. The economics of theatrical performance are great if you're operating eight shows a week in front of a thousand people at two hundred dollars a seat. They're not great if you're trying to recoup a hundred million dollar production budget. The math doesn't carry over. There's no version of "the movies" that is also "live theater," because the form of the movie has been "captured performance played back" since the medium was invented. That's the thing artificial intelligence can substitute for, and there's no live fallback to retreat to.
The Test Case Already Happened With Books
Look at what happened to fiction. Self-publishing platforms removed the gatekeepers. Then automated tools made it possible to produce passable books in volume. The result wasn't that human authors stopped writing. The result was that the middle of the market got flooded, the top of the market kept its name brand authors, and everyone else fought over the residue. Bestsellers still sell. The midlist got a lot harder. New entrants have a much harder time breaking through the noise.
Movies are heading there. The blockbuster will still be the blockbuster, because part of what you're buying is the cultural moment. The midlist movie, the sort of film that used to make twenty million dollars and let a director feed their family, is the part that gets squeezed. Streaming services already prefer machine generated thumbnails, machine written summaries, and algorithmically scored scripts. The next step is machine generated content of the same shape. That isn't speculative. The pipeline is being built right now.
Music doesn't follow that script because the live show is uncopyable. There's no equivalent to the live show in cinema, and that's the asymmetry that decides which industry has a real defense and which one doesn't.
What I Watch For
Three signals.
One. Watch the box office distribution, not the top line. The top ten will look healthy for a long time because Marvel and Disney and Pixar and the major franchises will keep coming. The interesting question is what happens at rank fifty through a hundred and fifty. The midlist movie. If those numbers compress, that's the squeeze.
Two. Watch what subscription streaming services do with their original budgets. If the spend on prestige series and tentpole films stays flat or grows but the spend on midbudget originals drops, that's the algorithm picking the same fight in a different venue.
Three. Watch live music ticket prices. They've been climbing for years and the trend has nothing to do with artificial intelligence. The fact that prices keep going up is the market telling you that the live experience has more pricing power, not less, in the era of infinite recorded supply. The same trend would apply to movies if movies had a live equivalent. They don't, which is the point.
The Take
Music and movies are usually grouped together because they both come out of the entertainment business. The supply structure makes them almost opposite cases. Music is high supply, low marginal cost, and the value has already migrated to the live experience that machines can't reproduce. Movies are low supply, high marginal cost, and the value is in the recording itself, which is exactly what machines are now able to produce.
The musicians I know are not happy about artificial intelligence, but they're not panicking either. They have a business that doesn't fully depend on the part the machines can take. The filmmakers I know are more worried, and the part of the industry they're most worried about is the part that already pays the bills for most working directors and writers, which is the middle. The headlines will keep being about the blockbusters. The damage is going to come from underneath them.