Infinite Content: Chapter 8

Case Studies in Media Disruption: Music, Print, Gaming, and Film/TV

This is the draft eighth chapter of my book, Infinite Content: AI, The Next Great Disruption of Media, and How to Navigate What’s Coming, due to be published by The MIT Press in 2026. The introductory chapter is available for free here. Subsequent draft chapters will be serialized for paid subscribers to The Mediator and can be found here.

---

You may think less of me for it, but I have a confession: I was a Napster user.

If you don’t remember, Napster was a peer-to-peer music file sharing service, launched in 1999. It wasn’t the first way to share files—technologies like Usenet and file transfer protocol (FTP) had been around for decades—but it was tailor made for music and user friendly. It indexed participating users’ hard drives on a centralized server, making it easy to find music, and then enabled you to download those songs directly from those users’ hard drives. (So, while it didn’t host pirated songs, it certainly encouraged piracy.)

Besides its dubious legality, Napster had other downsides. The files were relatively low fidelity (most songs were encoded at 128 kbps, less than 1/10 the fidelity of a CD) and many turned out to be corrupted, mislabeled, or incomplete. But two features drew me in: convenience and discovery. I used it to create a digital library of songs I already owned on CD, because searching and downloading was much quicker and easier than ripping CDs (which is legal). And because it was possible to see other users’ entire libraries, when I found someone who had a few of my favorite songs, I sampled and downloaded other songs I didn’t know. It was like an early version of finding someone’s playlist.

One night, my wife and I threw a party at our small apartment. I wired my computer to our stereo and ran the playlist from there. A friend of ours brought her new boyfriend to the party. Let’s call him Jeff, because I think his name was Jeff. Jeff took a keen interest in my setup. It turned out that he ran a business that built affinity graphs for music labels and radio stations. He would gather fans of, say, Britney Spears in a room and play them other similar music (like Christina Aguilera or Backstreet Boys) to see if they liked it.

“So, you download these songs even though they are pretty low quality, running them on those expensive speakers?”

“Yup.”

“And you have most of these on CD?”

“Yup.”

“And what is going on with this music?” He didn’t like that I had Led Zeppelin, Robert Cray, Manu Chao, Public Enemy, The Cure, Nirvana, Madonna, Stevie Wonder, Elton John, Simon and Garfunkel, Vivaldi, Rusted Root, Blues Traveler, Bob Marley, Aaliyah, Red Hot Chili Peppers, Andres Segovia, Black Sheep, Wilco, and Youssou N’Dour (among a lot of others) all on one hard drive. The premise of his business was that music fans stuck to one genre. He shook his head with a combination of bewilderment and mild disgust. “I hope you realize that this is not how people listen to music.”

Like much of the rest of the music industry at that time, Jeff didn’t understand disruption. He didn’t understand how a crappy product, like lo-fi, sometimes incomplete or mislabeled songs, might be “good enough” for a lot of consumers. He also didn’t understand that a new entrant might offer new features—like easier discovery, access to vast amounts of music, the ability to easily create playlists, etc.—that would change consumers’ definition of quality in music. (He had a vested interest in the status quo, which probably clouded his judgment.)

But that’s what happened. Consider again the chart I showed in Chapter 4 of global recorded music industry revenue (Figure 75). Little did he know it, but Jeff might have been shaking his head at my musical taste at the very moment that the music industry hit a peak that it wouldn’t regain for more than two decades. Recorded music industry revenue started declining in 2000 and would not re-achieve its 1999 level until 2021, 22 years later.

Figure 75. The Music Industry Went to Hell…And Back

Source: IFPI.

Recorded music experienced a very tough time from 1999 until around 2014-2015, but it has come back. As also shown in Figure 75, it grew every year since 2014 and in 2024, it reached almost $30 billion, up from $22 billion in 1999. On an absolute basis, that’s not so great—that’s just 1% compound annual growth over that period—and it’s down by about one-quarter when adjusted for inflation. But still. Let’s contrast that with the U.S. newspaper business.

U.S. newspaper revenue also started declining around 2000 (Figure 76). The difference is that it never came back. And it almost certainly never will.

Figure 76. The Newspaper Industry Went to Hell…And Stayed There

Source: News Media Alliance, formerly Newspaper Association of America (through 2012); Pew Research Center (2013-2020).

---

Let’s tie together the first half of the book and close out our long history lesson. Over the previous chapters, I’ve described how the internet disrupted media, leading to the tectonic trends that define the broader business today: stagnation of time spent, fragmentation of attention, price deflation, disintermediation of traditional intermediaries (and rise of creators), globalization, and concentration of power in a few platforms and attention in a few hits. These trends are evident across media. But disruption has affected each subsector differently.

Disruption is ultimately about how a certain type of new entrant—those with less performant, less expensive products—affects incumbents and how the incumbents respond, or fail to. In this chapter, we’ll survey the history of disruption for the key incumbents in music, print, gaming, and video: major music labels, U.S. newspapers, AAA video game publishers, and Hollywood studios, respectively. We’ll also explore the similarities and differences between the effects of disruption on each. Our goal is to tease out some of the nuances of the last disruption to sharpen our thinking about the implications of the next one.

It’s easy to get lost in the details as you read through these four histories, so here’s a preview of some important recurring patterns:

• Notice where value flows when something gets commoditized—to consumers who pay less for more, to complements that become more valuable, or to chokepoints that resolve bottlenecks in the value chain.

• Note also that gaming is the rare exception when disruption actually grows the market by attracting a lot of new customers.

• You’ll see that “quality moats”—technical barriers preventing upstarts from replicating incumbent quality—are important in determining how much disruption hurts.

• Note that consumer lock-in rarely saves incumbents in media, it only delays the inevitable.

• Notice how the consumer definition of quality shifts during media disruption, away from production values toward convenience, community, and authenticity.

• And finally, observe how the incumbents who best manage disruption have the governance structure necessary to make big, long-term bets.

These patterns will be important as we think about GenAI’s impact in the chapters ahead.

Besides using Napster, I have another admission. I found this chapter tough to write. A proper accounting of the disruptions of music, newspapers, gaming, and video could each be a book (and, in most cases, has been). Choosing the key inflection points and the right amount of exposition for readers with different levels of knowledge are both highly subjective. Hopefully, I hit the right high points and struck the right balance.