Ashlee Vance on robot fight documentaries, the brutal economics of streaming docs, and a China trip in the works
Feb 10, 2026 with Ashlee Vance
Key Points
- Netflix and other streamers now cap documentary budgets at roughly $2 million, down from years of overpaying, making it economically brutal to fund long-form critical journalism about tech companies.
- Core Memory is building a YouTube-native production model that keeps budgets lean and editorial control intact, while some established producers are already asking how to replicate the approach.
- Ashlee Vance is filming an ongoing robot fight documentary in San Francisco, planning to release it chapter-by-chapter as the sport evolves rather than selling footage to a single streamer.
Summary
Ashlee Vance is a journalist, author, and filmmaker who launched Core Memory roughly a year ago as an independent media company producing tech documentaries, YouTube content, and written journalism. The central tension in what he describes is straightforward: the documentary business is getting harder precisely as the stories worth telling are getting better.
The economics of streaming docs
The glory years of streaming are over for documentary makers. Vance says Netflix and most other streamers now cap documentary budgets at roughly $2 million, after a period when they were "overpaying for everything." Spread across one to six years of filming, that math is brutal. The problem compounds with consolidation — as tech companies acquire major media properties, editorial independence narrows. Vance flags a structural conflict: it's hard to sell a critical documentary about Apple to Netflix when those two companies are commercial rivals. The list of subjects worth following gets filtered through a web of platform allegiances before a single frame is shot.
YouTube as the escape valve
Vance argues YouTube still beats everything on distribution flexibility, and he thinks the industry is only in the early days of understanding what that means. His team built a lean production model at Bloomberg — Emmy-nominated work for a fraction of what Netflix or Hulu would spend — and has carried that into Core Memory. Once you sign with a streamer, he says, the budget bloats and the control disappears. He cites a telling example: Netflix line items for landlines running to $15,000, a relic of Hollywood's padding culture. Some established producers have already called Core Memory wanting to understand how the YouTube-native model works. Others haven't caught up.
Substack has been strong for text but hasn't solved the paywall problem for video. Vance acknowledges the tension without a clean answer — "you can't just give everything away" — and describes it as an open experiment across multiple distribution levers.
The robot fight documentary
Core Memory has been filming robot fight events in San Francisco for the past six to seven months, most recently at Rec's event at Kezar Pavilions. Rather than sell the footage to a streamer, Vance plans to release it as a living, ongoing documentary — chapter one covering the period so far, with more chapters as the sport evolves. He's also filmed what he describes as the only six-foot humanoid robot currently in the United States, though he withholds the backstory for now. The fights themselves are still more spectacle than sport, but he says they're improving fast, with six-foot-two robots expected within five months.
China and India on the horizon
Vance wants to do a China trip — humanoid robot factories, BYD, the works — but ran into visa difficulties during his Bloomberg years and is still working through the logistics. He mentions wanting to contact IShowSpeed to understand how a YouTuber visa compares to a journalist visa. An India episode is already in planning for this year.
The Elon read
Asked about SpaceX's pivot from Mars to the moon, Vance says he's still processing it. His read is that Elon is being unusually pragmatic: the US government's money is pointed at the moon, space data centers are creating new commercial logic, and the AI race is creating an urgency that's compressing his time horizon. The Mars vision, Vance argues, was always partly a recruitment religion — it made people want to work at SpaceX. Losing it is "a massive philosophical change." The cancellation of the Model S lands similarly for him: that car was the moment Tesla became real, and stopping production on it alongside the Mars pivot suggests Elon is all-in on AI and raising money to fund it, with SpaceX serving as the "sexy thing" that attracts capital for everything else.
Vance's own structural challenge mirrors what he describes in the industry: the best stories are long and uncertain, the platforms are consolidating, and the business model for serious independent journalism hasn't fully resolved. Core Memory is the bet that it can.