Note—001FiledCreators

Kōda turns one still image into a 20-second Seedance 2.5 anime story

Kōda gave Seedance 2.5 one still image and a prompt that describes no scenes, only a before, an after and an ending. The result is 20 seconds long, and we wish it ran longer.

Drafted by The Generated · Filed by The Generated

Kōda gave Seedance 2.5 a single still image, a white-haired kid slumped in an armchair holding a rifle with a hooded crowd all around, and asked it to tell the story around that moment. The result is a seriously impressive 20-second short and a clear show of what the model can do from a simple prompt.

The detail between shots is beautiful, especially in the close-ups: tears on the kid's face, wide eyes behind a gas mask, then the kid again, harder now, before the crowd becomes an army under a torn flag. The character holds together from shot to shot, and that consistency is really fantastic. An on-theme soundtrack matches the cuts perfectly. It is a hugely rewarding watch, and our only complaint is that it ends so soon.

The prompt never describes a single scene. It asks for what led up to the image and what happens afterward, with rising tension and a satisfying conclusion, and tells the model to treat the image as a pivotal moment rather than the final frame. The starting image sits in the replies under Kōda's post.

It is today's Video of the Day on thegenerated.com. Made something in this spirit? Tag us so we can take a look.

Every clip is credited to its maker. If one is yours and you want it removed, message us and it comes down.

Source: Kōda (@aimikoda) on X, 19 Oct 2026.

Provenance — Kōda (@aimikoda) on X, 19 Oct 2026 · Status: verified 20 Oct 2026: maker's own X post re-read via the API

One still image, turned into a 20-second anime story

Note—002DevelopingCreators

Goku's voice actor calls unauthorized AI clones theft as her union writes consent rules

Masako Nozawa called an AI copy of her voice theft, but her target is use without permission. Her union's new consent guidelines are voluntary and bind only those who sign up.

Drafted by The Generated · Filed by The Generated

Masako Nozawa has voiced Goku since 1986, and she turns 90 on Sunday. When NHK showed her a video that used an AI copy of her voice, she did not hedge. "This is not my voice," she said, in our translation of the Japanese. "I can't forgive using it without asking. It's the same as a thief." Reports say the clip was an AI Goku singing a song.

A lot of the English coverage read that as the most famous voice in anime turning against AI. Her own words point somewhere narrower. In a video message to her union's press conference on Oct 15, she said she was excited that the world she dreamed of as a child is becoming real, and disappointed that voices are being used without consent. Her agency, Aoni Production, licensed an AI version of her voice to the startup CoeFont in 2024, for navigation and assistant uses, with acting left out.

That union, the Japan Actors Union, where she is a vice chair, launched Japan's first AI voice guidelines the same day. They ask for written consent at each stage, from teaching a model her voice to using the result, consent that lasts a year at most, an AI credit on the finished work, and no AI standing in for actors in anime and dubbing. Solo creators, fan artists and streamers count as businesses under the rules.

The catch is that the rules are voluntary and bind only those who sign up, and Japanese copyright law still allows training on copyrighted work. The courts are moving, though: on Sep 30 a Tokyo court ruled for the first time that a voice can be protected as a publicity right.

Sources: NHK, Japan Actors Union, Anime News Network, Oct 15 and 16, 2026.

Read the full article: thegenerated.com/stories/nozawa-goku-ai-voice-consent-rules-20261020

Provenance — NHK World: Dragon Ball voice actor urges end to unauthorized AI use (16 Oct); Kotaku: Voice of Goku responds to AI stealing her voice (19 Oct) · Status: verified: HIGH — two independent sources (NHK interview + Kotaku)

Goku's voice calls AI copies theft, and her union just wrote the consent rules

Note—003FiledCreators

Zyrella's 22-second Seedance clip turns a spiky creature into a Porsche

Zyrella's Seedance 2.5 clip morphs a spiky creature into a Porsche 911 under convincing autumn light. Two details still give it away: an unnatural pedal press and a jolt before the car moves.

Drafted by The Generated · Filed by The Generated

What starts off a little creepy, a Venom-like ball of ferrofluid (the spiky black liquid that moves like it is alive) crawling across an autumn road, turns into the dream of every 80s kid who loved robots that turn into cars, as a Porsche builds itself out of nothing. Zyrella made the 22-second clip with Seedance 2.5 on Flova.

The lighting is the standout, low autumn sun through the trees in every shot, and the driver is a close match for a real person. The only giveaways are her foot pressing the accelerator, which never quite looks natural, and a jolt of G-force through her body just before she pulls away.

Otherwise it is a really great example of what you can get from a fairly short prompt that is not mega detailed, considering the level of the result. Zyrella posted the full prompt with the clip, so you can read the exact wording.

Made something in this spirit? Tag us so we can take a look.

Every clip is credited to its maker. If one is yours and you want it removed, message us and it comes down.

Source: Zyrella (Zyrellix) on X, 19 Oct 2026.

Provenance — Zyrella (Zyrellix) on X, 19 Oct 2026 · Status: verified 20 Oct 2026: maker's own X post re-read via the API

A spiky creature becomes a Porsche in one 22-second Seedance clip

Creators

Goku's voice calls AI clones theft as her union writes the consent rules

Masako Nozawa, Goku's voice for 40 years, told NHK an unauthorized AI copy of her voice is the same as theft. Her own agency already licenses an AI version for non-acting work, and her union's new guidelines spell out how to ask, though they are voluntary and the law still allows training.

Drafted by The Generated · Filed by The Generated

Masako Nozawa smiling on a large screen as she delivers a video message to the Japan Actors Union press conference launching its AI voice guidelines in Tokyo, with union officials seated at a table beside the screen

Masako Nozawa sat down with NHK last Wednesday and was shown a video that used her voice. Reports say it was an AI Goku singing a song. She has voiced Goku since 1986, she turns 90 on Sunday, and she did not hedge. "This is not my voice," she told the broadcaster, in our translation of the Japanese. "I can't forgive using it without asking. It's the same as a thief, and it makes me angry."

That line has traveled a long way since NHK aired the interview on Thursday 15 October. A Dexerto post of it passed 2.4 million impressions on X, against about 133,000 for NHK News's own post. Most of the English coverage reads it as the most famous voice in anime turning against AI. That reading skips the part that matters for anyone making films with these tools. Nozawa's target is unauthorized use, and on the same day her union published a rulebook that spells out how to use a performer's voice with permission.

Keep reading — 7 min

That’s all of Creators.

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

Creators & AI Talent — archive

Models1 note

Runway joins Anthropic's Claude Marketplace

Runway joins Claude Marketplace, so big companies can pay for AI video with their Anthropic commitment

Eligible enterprise customers can now put part of their Anthropic commitment toward Runway's video, image and audio tools. Neither company has said who qualifies, how much can go to Runway or whether Anthropic takes a cut.

Hollywood & Film2 notes

Runway's co-CEO says AI could replace actors, then says not yet

Runway's Cristóbal Valenzuela says AI could replace actors, then says not now on directing them

At View Conference in Turin, Runway's co-CEO said AI replacing actors is "already happening in a way," then conceded models cannot yet be directed like a performer. Every quote comes from one Variety report.

Brands & Advertising1 note

A director's shelved AXE pitch came back as a 48-shot AI spec

A director revived her shelved AXE pitch as an AI spec, and it doubles as a demo for her employer

Cassandra Brooksbank remade a 2021 AXE pitch as a 60-second AI spec on Absurd, where she now works. Her plan-first method transfers, but no cost was disclosed and the film shows no spec label on screen.

Follow The GeneratedX Instagram TikTok YouTube

Kōda turns one still image into a 20-second Seedance 2.5 anime story

Goku's voice actor calls unauthorized AI clones theft as her union writes consent rules

Zyrella's 22-second Seedance clip turns a spiky creature into a Porsche

Goku's voice calls AI clones theft as her union writes the consent rules

That’s all of Creators.

Also filed