The Central Cee Doja AI trend is a 13-second street-rap montage built from the shots of the Doja music video, and in the AI version someone else takes the lead role. On videosi you get that clip starring the person in your photo, with the trend sound already on it and no prompt to write. This guide answers the questions people ask after the basics.
Key takeaways
- The earliest AI recast we could date is from 17 September 2026, and clips were still being posted on 31 October 2026, so the trend is spreading.
- The sound is Doja by Central Cee, and the videosi video comes with it already on.
- One photo of one adult is enough; the masked crowd and the dogs stay as in the reference.
- A Standard 720p video starts at 10 credits, and failed renders return their credits.
- Label the post as AI-generated before you publish.
In this guide
- What is the Central Cee Doja AI trend, exactly?
- When did it start, and is it still going?
- How do I make the Doja video from one photo?
- Which photo works best?
- Who else appears in the video, and are they me?
- What song is used, and will it play on my video?
- Which AI model makes it, and can I use other tools?
- How much does it cost?
- Why did mine come out wrong?
- What should I know before posting it?
What is the Central Cee Doja AI trend, exactly?
It is a 13-second montage that follows the shots of the Doja music video: a low look up a grey tower block, a balcony ledge, a close-up in a round spotlight, two large dogs on leashes, a crowd of masked men in black, a car window and a courtyard dance. In the AI version the lead role goes to someone else, whether a character, a celebrity or the person in your photo. The structure stays fixed, which is why the clips look like one family. Nothing in this trend is filmed by the person posting it: the shots come from a reference clip and a video model swaps the lead.
Many remakes keep the black-clad crowd and the dogs exactly as they are, because those are part of what viewers recognise. Only the face and costume of the lead change.
When did it start, and is it still going?
The earliest AI recast we could date is from 17 September 2026 on TikTok, with a second wave in early October. A Vietnamese remake posted on 1 October 2026 passed three million views, and new clips were still appearing on 31 October 2026. That puts the trend in a spreading stage as of November 2026, not a finished one. The song itself is older than the trend, so the music is not what is new here; the recasting is.
We have not confirmed a single original remake, so we do not name one. Dates here come from the earliest posts we could find, not from a platform report.
How do I make the Doja video from one photo?
Open the Central Cee Doja AI page, upload one clear photo of yourself, pick 16:9 or 9:16 and press Generate. The 13-second reference clip and the prompt are built in. Rendering takes several minutes, and you can close the page because the result is saved in My creations. You write nothing.
The format matters for framing. 16:9 keeps the wide look of the original; 9:16 crops it for TikTok, Reels and Shorts.
Which photo works best?
A sharp photo of one adult looking at the camera, with face, hair and shoulders in view and even light. The lead is shown from the waist up in most shots, so the model needs more than a close crop of a face. Plain, fitted clothes hold steady across the cuts, while busy patterns and loose layers tend to drift. Skip sunglasses and hats, because they hide the features the model uses to keep you recognisable in seven different shots.
What you wear in the photo is what you wear in the video. The uniform from the reference is removed on purpose.
Who else appears in the video, and are they me?
No. The lead is the person from your photo in every shot where the officer appears, including the car-window shot where the reference shows a different, curly-haired man. The masked men, the black-clad dancer and the two dogs stay as in the reference clip, with faces covered. If you want a group, that is a different kind of clip and needs several photos, which this page does not use.
This is a limit worth knowing before you pay: the crowd is scenery, not a cast you can fill with friends.
What song is used, and will it play on my video?
The song is Doja by Central Cee. TikTok lists the sound under that name on most remakes and lyric edits. The finished videosi video already carries the trend sound, lined up with the 13 seconds, so you do not have to add it. The lead does not lip-sync the lyrics, which is a common surprise. You can hear the original on its Wikipedia page.
Posting a video with a copyrighted song is your decision. Platforms can mute or limit it, so check the rules of the app you post to.
Which AI model makes it, and can I use other tools?
videosi uses ByteDance's Seedance video model, with Seedance 2.0 Mini as the default and faster or higher-quality tiers one click away. The prompt has five parts: replace the lead, keep the footage, lock the identity, follow a timeline and set rules. We do not publish the full prompt. You can try other tools, but expect several attempts, because the dogs, the crowd and the cuts drift without a reference clip.
The reference clip is what keeps the shot order. Without it, models tend to invent one generic rapper and drop the spotlight and the dogs.
How much does it cost?
A Standard 720p Central Cee Doja video starts at 10 credits, and higher models and resolutions cost more. Credit packs start at $6.99 for 10 credits and never expire. There is no free generation. If a render fails, its credits return automatically, so you do not pay for a clip that never arrives.
Prices change, so check the live price in the generator before you confirm.
Why did mine come out wrong?
The common causes are a photo that hides the face, a group shot, a profile angle or heavy accessories. Several shots are wide or fast, so likeness in those depends on hair, build and clothes more than on the face. The close-up in the round spotlight is where your face is clearest. Likeness is not guaranteed on any render, and a failed render returns its credits.
If the lead looks like someone else, try again with a front-facing photo that shows your upper body in even light.
What should I know before posting it?
Label the post as AI-generated where the platform asks for it. Use only your own photo or that of an adult who has agreed, because videosi does not accept celebrities, public figures or minors. The clip is a re-staging for fun and is not connected to Central Cee, his label or the makers of the music video. Do not present it as an official video.
Photos are deleted within 30 days, and the finished MP4 downloads without a watermark.
Where to go next
If you only want the quick version, the trend page has the beat table, the fast facts and the short answers. This guide exists for the questions that come after: what stays fixed, what you control and what to expect from the result. When more people post remakes, the page and this guide will be updated with new dates.
