Skip to content
Note—003DevelopingFilm

Nigeria's 'first Yoruba photorealistic AI feature' premieres online

A Yoruba-language AI feature sold its own tickets on the Docuth rental app and the premiere stalled on sign-ups, payments and playback. The studio blames a viewer surge but has published no viewer or sales numbers.

Drafted by The Generated · Filed by The Generated

Asiri Ilu Awon Osu, a two-part, three-hour Yoruba-language feature from Ibadan's Mysterious Tech Studio and actress and producer Tanwa Hameed, premiered on the Docuth rental app on Monday, Oct 5. By Tuesday the studio was apologizing to viewers at home and abroad who could not register, pay or, in some cases, watch a film they had already bought.

The studio calls it a surge of viewers, but nobody has published a number to back that, not the studio, not Docuth and not The Punch, which carried the apology. The public counters are small: Docuth's Android app showed 1K+ installs on Google Play on Thursday, and the first official trailer has 413 YouTube views. Sales through WhatsApp groups and Facebook pages would not show up there, so a real rush is possible, though a young app does not need a stampede to buckle.

The makers say about 80% of the visuals were generated or assisted by AI, and they have never named a tool. The trailers show a genuine hybrid, with generated war and sci-fi set pieces beside live-action drama.

In practice, the generation held up and the checkout did not. A small team can now make a feature for a language market the big streamers rarely commission for and sell it at its own price, so before a pay-per-view premiere, ask the platform for its peak sign-up, payment and streaming load, and test the purchase path from a fresh phone in every market you sell to.

Source: The Punch, Oct 8, 2026; Google Play and YouTube, read Oct 8, 2026.

Provenance — The Punch — Streaming glitches disrupt Yoruba AI film premiere; VP Land — Nigerian team completes a three-hour Yoruba feature with AI; Vanguard — milestone despite Guinness rejection · Status: verified: MEDIUM (The Punch on the premiere; film covered since Aug by NTA, Vanguard, VP Land)
A Yoruba AI feature sold its own tickets, and the app couldn't keep up
LeadNote—001DevelopingFilm

BBC staff push back after the director-general calls an AI-made 'Doctor Who' 'pretty good'

BBC staff asked director-general Matt Brittin whether the BBC should fully reject generative AI after he praised an AI Doctor Who episode. He clarified that humans are leading but did not retract the remark, and BBC policy has not changed.

Drafted by The Generated · Filed by The Generated

At a BBC staff town hall on Wednesday, Oct 7, employees challenged director-general Matt Brittin over his Sept 29 remark that a 50-minute AI-generated Doctor Who episode someone sent him was "pretty good." One question, upvoted 86 times, pointed out that Doctor Who is made in Cardiff and said young people "do not want this" content. Another asked whether it was time for the BBC to fully reject generative AI.

Brittin joked that he "got in trouble" but did not retract the remark. He said he was making a point about the pace of change, that he wants to speed up testing and learning, and that "humans are leading." He has not named the episode or who made it.

The timing matters because the BBC is running a competitive tender for the next producer of Doctor Who. Its own supplier guidance already tells producers to confirm any proposed AI use with their commissioner, and to disclose it directly wherever generated output could mislead an audience.

In practice, if a pitch to a UK broadcaster leans on generated footage, clear it with the commissioner before the deck goes in and plan the disclosure line from day one.

Source: Deadline, Oct 8, 2026; BBC AI transparency guidance.

Provenance — Deadline; Gizmodo: BBC boss walks back Doctor Who AI comments; Bleeding Cool · Status: verified: HIGH (Deadline + Gizmodo + Bleeding Cool)
BBC staff challenge their director-general over his 'pretty good' AI Doctor Who
Note—002DevelopingFilm

Skydance's Day 1 memo: 'technology must serve the art'

Skydance's first staff memo says AI must serve the art, but the combined studio starts with almost $80 billion of debt and no AI production policy yet.

Drafted by The Generated · Filed by The Generated

Skydance, the company formed when Paramount Skydance closed its Warner Bros. Discovery deal on 6 October, used its first staff memo to set an AI rule. David Ellison and Ynon Kreiz wrote that the studio will embrace AI "to make our businesses even more productive," then added the line crews will quote back at them: "technology must serve the art, never the other way around."

The money points the other way. The company starts with almost $80 billion of debt, S&P puts its leverage above seven times earnings according to the FT, and the merger release targets $6 billion-plus in savings over three years, with technology named first. The Ankler reports investors are already pricing Skydance against the AI market through Larry Ellison's Oracle stake, even though Skydance has no data center exposure of its own.

No AI production policy exists yet, and the memo names no tool, pipeline or role. Our read is that a studio carrying that much debt will buy AI that saves money first. In practice, if you build or sell AI tools for film, pitch what artists can make with them rather than hours saved, because the first Skydance pipeline deal will show which half of the memo wins.

Source: Skydance Day 1 memo via Variety, Skydance release, FT, The Ankler, 6 Oct 2026.

Provenance — THR — Skydance's $80 Billion Debt; The Ankler — Skydance Tightrope: Debt Tied to AI Economy; Variety — Ellison's Day One press conference; Cartoon Brew — Skydance finalizes merger · Status: verified: MEDIUM — merger close confirmed by Variety, Cartoon Brew and THR. The memo line is THR only; the AI-debt frami
Hollywood's newest giant says AI serves the art. Its debt gets a vote.
Film

Fox Nation's "all-AI" Houdini series runs on actors wearing new faces

Fox Nation premieres the three-part Houdini: Back from the Dead on 14 October, billed as its first production made entirely with AI imagery. Red Rock Films says actors performed the roles and AI changed their faces, built on licensed archive. The tools are unnamed and the series is not the first of its kind.

Drafted by The Generated · Filed by The Generated
Side by side: an archive photo of Harry Houdini hanging upside down over his Water Torture Cell, and a color trailer frame of an AI-altered Houdini upside down in water, labeled AI GENERATED

The trailer runs 30 seconds, and every frame of it carries the same small white label: AI GENERATED. A man in a dark suit stands alone on a stage, a woman in a beaded 1920s dress watches from the stalls, a hand claws up out of the dirt, and a man in chains stares into the lens. Near the end, Harry Houdini hangs upside down in his Water Torture Cell, hair floating, eyes open. Over the top of it, a voice announces: "Through the magic of artificial intelligence, I will return to tell you my story in my own words. I'm Harry Houdini."

Fox Nation, the Fox News streaming service, premieres Houdini: Back from the Dead on Wednesday 14 October, two and a half weeks before the 100th anniversary of the magician's death on 31 October 1926. Its announcement on Thursday billed the three-part series as the platform's first full-length, multi-episode production "to rely entirely on AI-generated imagery". The series was produced by Red Rock Films, the Emmy-winning natural-history house behind Secrets of the Whales and Penguin Town, through its AI division, Red Rock AI. Read the release closely, though, and the most interesting thing about the show is how much of it is human.

Keep reading — 7 min

That’s all of Film.

9 Oct · 3 notes, filed as the record moved.

Hollywood & Film — archive

Also filed

Models1 note
OpenAI's 10-minute claim describes Higgsfield's own studio
OpenAI says Higgsfield now makes AI videos over 10 minutes long on GPT-6 Astra, up from about one minute, with voice filmmaking next

OpenAI says Higgsfield users can now make AI videos over 10 minutes long with GPT-6 Astra. Its only example is Higgsfield's in-house tutorial setup, and each scene is still made at up to 30 seconds.

Harry Potter, out of the wizarding world and into the dojo
nachos2d: #malfoid Malfoid’s Diary | Ep. 2

nachos2d's 30-second anime short sends Harry Potter to a karate dojo, where every hit pops a little heart out of his mouth. The post names no tools, so neither do we.

Disney bids $20M for Character.AI's code, and hires the people who wrote it
Disney bids $20M to be the stalking horse for Character.AI's assets as the chatbot startup files for Chapter 11

Disney bid $20 million to license some of bankrupt Character.AI's code, while most of the startup's staff joined Disney. The app, its users and their chats are not part of the deal, and rival bids can still top it.