The gas station dance AI trend is a 12-second night clip in which a person pushes a car door open at a gas station, hops out and breaks into a bouncing, low-crouch dance. On videosi you make it from one photo of yourself, pick a car, press Generate and download a silent MP4, with no prompt to write. The original clip is a man dressed as a cartoon rich kid, and the AI version borrows only the moves and the timing.
Key takeaways
- The character in the original is Bling-Bling Boy from Johnny Test, a wealthy kid, which is why the outfit is white and gold and the car is gold.
- Templates and bare prompts fail for different reasons: templates expire, and prompts cannot carry a motion reference.
- Likeness drifts most when you turn your back or spin, so the photo matters more than the car.
- A standard 12-second video (Standard model, 720p) costs 10 credits, it is not free, and a failed render returns its credits.
- Label the result as AI-generated where you post, and only use a photo of an adult who agreed.
In this guide
- Who is the character in the original gas station dance video?
- What does "gas station guy bopping" or "mop dance" mean?
- Why does a gas station dance template stop working?
- Why is a viral gas station dance prompt not enough?
- Why does my face change after the turn?
- Which car and outfit should I pick?
- Is a dance in the parking lot AI video the same thing?
- What should I do before I post it?
- Can I use someone else's photo?
- Can I make it in Kling, ChatGPT or Sora?
- What does it cost, and what if a render fails?
Who is the character in the original gas station dance video?
The clip that the gas station dance AI page copies shows a man in a white and gold outfit, dressed as Bling-Bling Boy from the cartoon Johnny Test, stepping out of a gold car at a night gas station. Wikipedia's page on the series describes Eugene "Bling-Bling Boy" Hamilton as a wealthy technological rival who holds an unrequited crush on Susan. That explains the styling: a rich kid arrives in a gold car, so the costume and the car both shout money.
Reposts of the clip were on TikTok by 8 October 2026, credited to an Instagram account. We have not confirmed who first made it. What matters for an AI remake is the door, the exit, the clap, the spin, the crouch and the dip. videosi copies those 12 seconds of movement and nothing of the man's face, hair or costume.
What does "gas station guy bopping" or "mop dance" mean?
They are captions, not different videos. People searching for the gas station guy bopping, the gas station guy mopping or the mop dance are almost always describing the same short night clip. The low crouch with both arms pushed forward and the hands together, rocking side to side, is the move everyone copies; where the nickname "mop dance" comes from is unconfirmed.
Nothing needs configuring for any of those names.
Why does a gas station dance template stop working?
Because a template is a shortcut owned by someone else. A template account can change it, remove it or replace it with a newer version, and the template only works while it stays up in the editor you are using. From about 16 October 2026, remakes made with CapCut templates were being posted with the caption "bro turned the gas station into a whole dance floor", and templates like that are exactly the kind that come and go. Check inside CapCut for what is available today; videosi is independent and not affiliated with TikTok or CapCut.
The videosi version is a web tool instead. You upload a photo, choose 1 of 4 cars and a format, and download a silent MP4 that you can use in any editor, which means the result does not depend on a template staying online.
Why is a viral gas station dance prompt not enough?
A prompt describes a scene in words, but this trend is a specific sequence of movements, and words are a poor way to pin down timing. Without a motion reference, most models make a generic dance in a parking lot. The door opening, the exit, the clap with your back turned and the crouching groove are what make it the gas station dance, and they are the first things to disappear.
The videosi version pairs a five-part prompt with a 12-second motion reference clip and runs it on ByteDance's Seedance model. Even the opening line shows the intent: "Photorealistic live-action phone footage, 12 seconds, one continuous handheld shot at night at a gas station, filmed by a friend standing a few steps from the car." The parts after it handle identity, motion, scene, timeline and rules. We do not publish the full text.
Why does my face change after the turn?
Look at the beats. Between about 3 and 5 seconds you have your back to the camera and clap above your head, then between 5 and 6.5 seconds you spin round. For a stretch of the clip the model cannot see the face it is supposed to keep, and when it reappears the model may fill it in from habit instead of from your photo. That is why likeness drifts most right after the turn, and why it is never guaranteed.
You cannot remove this, but you can reduce it. The recipe puts the identity lock first, so your photo is the only source for the face, hair, build, outfit and shoes. On your side, keep the face clear: no sunglasses, masks or caps pulled low. Use a photo of just you, because a second face can end up on the dancer. And expect that a fast spin or the deep dip around the 9 second mark may blur a hand or slide a foot, so a second try is normal.
Which car and outfit should I pick?
There are 4 cars. The silver sedan is the default and the closest to most remakes of the trend. The dark red sedan and the bright blue muscle sedan are bolder choices, and the black car is a matte muscle car whose back seat is visibly piled with bundles of cash. The car sets the mood of the first seconds, when the camera is watching the open door, so pick the one that matches the joke you want.
The outfit is not chosen in the app at all: it comes from your photo. A photo that shows your whole body, legs and shoes included, gives the model your real clothes. A face crop leaves it to invent trousers and sneakers. Most remakes wear something bold, such as a puffer jacket, a white hoodie, a suit or a costume, and bright or light colours read well under the white canopy lights at night.
Is a dance in the parking lot AI video the same thing?
No, and mixing them up sends people to the wrong tool. The sunset parking lot dance is a separate clip of three friends from Russia, with one dancing, one filming a selfie and one getting out of a car, which spread in late September 2026. It is sometimes also called the gas station dance, but it is a three-person daytime scene. The gas station dance AI page makes the one-person night version only: one dancer, one car, one handheld shot.
So if you search for a dance in the parking lot AI video and want three people at sunset, this is not the page. If you want to be the single person stepping out of the car under a gas station canopy, it is.
What should I do before I post it?
Label it as AI-generated. The clip is made to look like real phone footage, and platforms ask you to say when realistic content is synthetic. YouTube's help page says creators must disclose AI-altered content that looks realistic, such as a scene that never happened, using the "AI use" setting in YouTube Studio, and says creators who consistently fail to disclose can face penalties. Other apps have their own labelling tools, so check the one you post from before you upload.
Then add sound. The videosi result is silent, so pick the sound you heard on the trend from the app's own library when you post. The song is unconfirmed: the reposts we checked use original sounds, so we do not name one.
Can I use someone else's photo?
Only the photo of an adult who has agreed to it. videosi does not accept photos of celebrities, public figures or minors, and users must be 18 or older. The clip puts a real-looking person in a video they never filmed, so it needs their consent. Use a photo of one person, since group photos can mix faces, and remember that photos and results are deleted within 30 days.
Can I make it in Kling, ChatGPT or Sora?
You can try, with different odds. Some tutorials use Kling's recreate feature together with the trend clip, which gives it a motion reference. Chat apps like ChatGPT do not take a motion reference, so they tend to produce a generic dance. We have not tested every tool, and tools change quickly, so treat this as a direction rather than a promise.
What the gas station dance AI page adds is the pieces already put together: the Seedance model, the motion reference, the five-part prompt and the identity lock.
What does it cost, and what if a render fails?
It is not free, and there is no free trial. A standard 12-second video (Standard model, 720p) costs 10 credits. The smallest pack is $6.99 for 10 credits, which never expire, and plans start at $29.90 a month. A failed, timed-out or blocked render returns its credits automatically. Rendering takes several minutes, and getting the 12 seconds to land beat for beat can take more than one try, so budget for a retry if you are fussy about the spin and the dip.
