Eventually, the sample ate the song.

In the late 80s, a producer would loop four bars of a James Brown record and call it a beat. The label would call it theft. The clearance economy got built around it (at De La Soul’s expense), the publishing splits got renegotiated, and within a decade, the sample had eaten the song. The Amen break became one of the most-used drum loops in modern music. The Funky Drummer became a cultural building block. The originators didn’t make enough money from that. But the derivatives often wound up becoming bigger than the source material.

I keep thinking about that watching what's happening across podcast and creator media. Charlie Warzel and Ed Elson framed it as "the clip economy" on Galaxy Brain this month — short clips pulled from podcasts and livestreams have become the dominant unit of online media, outperforming the long-form they were carved from rather than promoting it.

But let’s get real. Aren’t movie trailers just clipped films, and entertainment content themselves? Aren’t GIFs just clipped reaction shots that could never be made in a lab? Aren’t samples just clipped songs that have been manipulated to create something 100% new? Every medium eventually arrives at the same place. The snippet that started as promotion ends up as the product. It’s inverted thinking that his hard to learn if you aren’t prepared to unlearn everything you’ve been taught about content.

The cleanest proof point is sitting on OpenAI's books right now. The company paid roughly $200M earlier this month for TBPN, a daily tech show that averages around 7,000 live viewers per episode. That number would make the deal absurd if you valued it on the show. Value it on the clips — average clip viewership of about 257,000, ads baked directly into the short-form, $5M in 2025 revenue tracking to $30M in 2026 — and the math gets coherent fast. The platforms have already noticed. Kick runs an in-house clipping program. MrBeast launched his own clipping platform to seed Beast Land theme park content into every algorithm at once. Both presidential campaigns in 2024 paid clippers to flood feeds. Whoever owns the clip distribution layer owns the modern equivalent of MTV rotation, and the price tag for that real estate is going up by the week.

The single biggest strategic shift in creator media this year is that the smart ones have learned to mine their own catalog before someone else does. Clyde Stubblefield is the cautionary tale. The drummer behind the Funky Drummer break — sampled on more than 1,300 songs, the rhythmic spine of a generation of hip-hop — was paid a session musician's fee in 1969 and never received songwriter credit. By the time he tried to capture some of the upside, releasing licensable versions of his own beats through the Copyright Criminals project, the value had already been mined out of him by everyone else. He died in 2017, having watched other people get rich off twenty seconds of his brilliance.

Today's creators have read that history. Andrew Tate built the original blueprint for getting in front of it, and the mechanics are worth studying even if the politics make you wince. He paid subscribers a commission to clip his livestreams and post them with affiliate links, and by the time TikTok banned him his clips had been viewed more than 11 billion times. The streamer N3on now reportedly pays his clipping army a combined million dollars a month, with top earners pulling seven figures a year. MrBeast built his own clipping platform rather than let other people own the cuts of Beast Land. The unifying logic is what I’d call a reverse Stubblefield — capture the value of your own material before the ecosystem does it for you, and turn the long-form into the publishing catalog you actually own the splits on. If you're sitting on long-form and treating clips as marketing exhaust someone else can monetize, you're auditioning for the session musician role.

The brand side of this is where the opportunity is most underbuilt, and it's the part that should keep CMOs awake. Brands are still buying long-form sponsorships the way they bought 30-second pre-rolls a decade ago. They commission a piece of content, they get a host-read or a brand integration, they cross their fingers and hope the YouTube thumbnail does some work, and the clips just happen to them. TBPN inverted that. They bake "this clip was brought to you by..." tags directly into their short-form output, which is part of how a show with 7,000 live viewers built a $30M revenue run rate. The unit economics of the brand integration travel with the clip, not the episode. [Prof G Media] The $200M check OpenAI wrote bought a comms apparatus that produces hundreds of clipped, branded impressions a day, in formats the algorithms already love.

The implication is that the next brand-funded production should be designed from frame one to shed 40 short-form pieces a week, with the brand wired into the cuttable moments instead of bolted onto the episode container. That changes the brief, the budget, and the talent. The traditional content brief asks for a topic and a runtime. The clip-economy brief asks for the ten cuttable moments inside the episode — the demo, the reveal, the disagreement, the visual, the line — and treats the long-form as connective tissue. If your producers can't name the moments before they shoot, you're commissioning marketing exhaust. Media dollars and modern social media platforms were built for (and their algorithms encourage) this.

It also changes who you pay. Clippers earning $20–30K a month with top performers pulling seven figures are operating as a distribution channel, and they should be funded like one. The MrBeast model of running your own clip platform is the extreme version. The minimum viable version is a revenue share with a half dozen clippers who already know your audience, written into the deal alongside the talent fee. And it changes how the long-form gets shot. Most brand-funded productions still get made the way a network sitcom was shot in 2008 — wide masters, slow setups, dialogue that needs context to land. None of that clips. The brands winning right now are commissioning shows dense with self-contained moments by design: tight framing, strong visuals, lines that work without setup, graphics built to survive being cropped to 9:16. The long-form still has to be good, but goodness now includes how much it sheds. And media dollars fuel their distribution.

The sample ate the song forty years ago, and the music industry spent a decade pretending it hadn't. Long-form media has the same window now, and it's going to be shorter.

Until next week, Ian Schafer, Ensemble, and the POST CREDITS team

— Ian

PANEL: The Future of Entertainment

I’ll be in-conversation with Scalable Pod’s Jasmine Enberg, along with Rich Bloom, GM Creator Programs and EVP, Business Development, Tubi, and Content Creator, Actress and Comedian, Amanda McCants.