My Vampire System runs 4,192 episodes. It launched four years ago and has been streamed more than 1.5 billion times. To hear all of it, you either burn through 30 minutes of daily free listening for four years, or you pay per episode — anywhere from a few cents to over $3 each. Some users have spent hundreds, even thousands, of dollars on a single series.
Behind the show stands no professional production team. It comes from Pocket FM, an India-headquartered audio entertainment platform founded in Bangalore in 2018 by three friends in their mid-twenties. The platform makes money almost entirely through microtransactions — there is no subscription. This one series alone has pulled in an estimated $90 million in revenue.
The more revealing number is the cost structure. In 2024, Pocket FM's content cost roughly $2,000 per hour of finished audio, and the company still lost nearly $20 million on operations. Then it terminated contracts with 200 writers and shifted fully to an AI-powered user-generated content model. The cost per hour dropped to $30. Over the past 12 months, revenue reached about $400 million — more than double the prior year — and the company turned cash-flow positive, with roughly $10 million in profit.
Most coverage treats this as a success story. That misses the point. The real question is colder: when the marginal cost of content production falls to $30 an hour, how much of the entertainment industry's underlying logic still has anything to do with "creation"?
1. The End of the Hit Lottery
Traditional entertainment's cost structure means every piece of content is a heavy bet placed before release. A series starts at millions of dollars and takes two years from script to screen, with the hit-or-miss wager concentrated in a handful of professional teams. The moat of that system is not creativity — it is filtering. Professional channels act as the funnel; only projects that clear layers of greenlighting ever reach an audience.
Pocket FM moved the entire funnel to the audience. CEO Rohan Nayak puts it bluntly: "I used to wonder why it was so hard to build a hit. Now I don't care about the hit rate anymore. Even if the hit rate of user-generated content is lower than professionally produced content, the absolute number of hits is more than ten times higher."
That sentence deserves a second read. The decision variable has changed. The traditional model optimizes the hit rate; Pocket FM optimizes the denominator. When production is cheap enough, individual failures stop mattering — only volume does. In August alone, more than 200,000 hours of new content was uploaded to the platform. Since it opened AI writing tools to all users in early 2025, more than 550,000 creators have published work there.
This is not efficiency improvement. It is a migration of the filtering mechanism. Professional editorial judgment has been replaced by two things: AI, which drives the production threshold to nearly zero, and pay-per-episode microtransactions, which filter with real money. Hits are no longer predicted. They are grown.
2. The Hollywood Control Group
Incumbents are not sitting still. DramaBox joined the Disney Accelerator in 2025. Fox Entertainment invested in Ukraine's Holywater. Sony, Paramount's BET, and Versant all have vertical microdramas in production. Holywater co-CEO Bogdan Nesvit is explicit about the logic: microdramas are essentially IP incubators, and the platforms' minute-level behavioral data directly informs what to develop next.
But the contrast remains stark. Hollywood's path is "professional production plus data assistance," still showing reverence for production costs. Pocket FM's path is "user production plus AI supply," with production costs compressed toward zero. Hollywood designs its hits; Pocket FM's hits are survivorship-bias outcomes — and that is exactly what Nayak wants.
The user data is even more uncomfortable for industry orthodoxy. Brandan Dennehy, a 20-year Hollywood veteran who built Pocket FM's US content operation, observed: "A lot of people genuinely don't care whether the production is professional. Once content is pushed in front of users and grabs their attention, all the inherited assumptions about what entertainment is supposed to look like just evaporate." The less he applied traditional Hollywood storytelling, the better his shows performed.
This punctures a default assumption: audiences are not rejecting low production values — they are rejecting boredom. Payment behavior has already voted. The platform now has more than 2.5 million paying users averaging 150 minutes of daily use, and roughly 80% of revenue comes from the US market.
3. The Volume Game and Filter Migration
Distill Pocket FM's structure into a reusable framework and you get something I would call filter migration: when a category's production cost falls past a threshold, the industry's core competitive activity stops happening on the production side and moves to distribution and filtering.
The framework has three layers. Production: AI pushes marginal cost to near zero and supply becomes effectively unlimited. Filtering: market mechanisms — paywall unlock rates, retention data — replace professional pre-approval as the only signal that matters. Moat: competitive advantage shifts from "who can make good content" to "who controls the distribution surface and the behavioral data."
Nayak's own business logic is a footnote to the framework: "Give me 100,000 good series, and I'll deliver $5 billion in annual recurring revenue." Notice what is absent from that equation — hit prediction. Only scale.
None of this is new in kind, only in reach. Open-source software pushed the cost of code toward zero, and filtering migrated to stars and download counts. Short video pushed the cost of video content toward zero, and filtering migrated to recommendation algorithms. Pocket FM simply carried the same logic into long-form audio storytelling — a category traditionally assumed to be the most production-intensive of all.
4. Which Industries Are Next in Line
Apply the framework elsewhere and you can start judging which "filter migrations" are imminent.
Indie game development: generative tools are already compressing the cost of art, audio, and level design. Once prototype costs fall to Pocket FM levels, games will see the same volume game — thousands of titles shipped, paywalls doing the filtering, publisher gatekeeping quietly devalued.
Professional document markets: legal contract templates, industry reports, courseware. These categories are still guarded on the production side by professional institutions, but AI already produces above the passing grade. When users can get a serviceable draft for cents per call, the pricing logic of paid reports and courses gets restructured from underneath.
AI video itself: Pocket FM debated a full pivot to video back in 2024 but shelved it over retention and cost. This year it quietly launched Pocket Saga, already clearing $1 million in monthly revenue. Nayak's stated plan is to convert the platform's 200,000 hours of audio into AI video — "a massive undertaking, but that day will definitely come." When video's marginal cost also crosses the threshold, the blast radius stops being audio and becomes the entire content pyramid.
One sober caveat: the volume game does not eliminate professional content. It evicts professional content from the middle and confines it to the premium tip of the pyramid. Netflix-style prestige production will survive — but it will lose its status as the default way content reaches audiences, the way handmade furniture remains expensive yet furnishes nobody's world.
5. What to Do About It
If you run a content platform: audit your cost structure now. When a competitor's marginal cost is an order of magnitude below yours, everything you spend on editorial curation becomes a fixed-cost liability. Move your filtering from pre-approval to market validation before someone else does it for you.
If you are a professional creator: do not compete with the volume game on supply. Your leverage sits after the migration — brand, cross-media IP operations, and trust itself. The gaming and short-video playbooks agree on this: hits emerge from UGC, but monetization at the top comes from professionalizing the original IP.
If you are an investor: stop evaluating AI content platforms by hit rate. Watch two metrics instead — whether marginal cost per unit has crossed the threshold, and whether the user behavioral data compound into exclusivity. The first decides whether the model works; the second decides who owns the moat.
If you are a traditional media or film company: treat AI production as an IP test bed, not a substitute. Holywater's open-data model is the template — use cheap content to probe demand, then decide where the expensive bets go.
The next decade of the content industry will not be won by whoever makes the best content. It will be won by whoever owns the most accurate filtering machine. Production is becoming free. Judgment is becoming expensive.