Gangnam Style AI: put your trio in the spotlight
Gangnam Style AI turns three photos into a 19-second night-time selfie video with the original sound. One person sings into the camera, the second dances a chest-and-hip solo, the third spins in and points at the lens, then all three shout the hook together. Upload your photos, press one button, and a standard video costs $9.99.
- 3 photos
- 19 seconds
- 16:9 widescreen
- Original sound
- From $9.99
Cast the three spotlights
One clear photo of one person for each. All three are needed.
Photo 1
Photo 2
Photo 3
100 credits = one standard video = from $9.99. Credits for a failed video are returned.
How to make a Gangnam Style AI video
Three steps and a few minutes. There is no prompt to write and nothing to edit: the three faces are the only thing you choose.
1Add the three photos
Drop a portrait on each of the three spotlights. Photo 1 is the selfie holder in the centre, Photo 2 dances on the left and Photo 3 spins in from the right. All three are needed.
2Choose Standard or HD and confirm
Standard is 480p and costs 100 credits, HD is 720p and costs 300. Tick the box to confirm you may use all three photos, then press Make my video. Sign in when asked; your photos stay on the page.
3Wait, watch and download
Both pieces render at the same time and are joined with the original sound under them. When the player appears, watch it and download the 16:9 MP4. You can close the tab; the video waits in your account.
What is the Gangnam Style AI trend?
Gangnam Style AI is a short video format in which three people share one phone camera on an empty plaza at night. The person holding the phone grins, gasps and sings straight into the lens. The camera then swings to a second dancer, who sways with a hand on the chest and a hand on the hip, and to a third, who spins in, flips their hair and points at the viewer. In the last seconds all three crowd into the frame, open their mouths for the hook and throw their hands in the air.
The sound is the chorus of PSY's “Gangnam Style”, released on 15 July 2012 and named for the wealthy Seoul district it gently mocks. The song's video became the first on YouTube to pass one billion views, on 21 December 2012, thanks to the invisible horse ride. But the clip behind this trend does not copy the horse dance. Its joke is the camera: whoever the lens lands on becomes the star of the whole song for a few seconds.
The version this site uses was found on TikTok in a post by @clipnova44.ai, and is credited here; the original video is not hosted on this site. Reports of the format point to a night-time clip shot in Seoul's Gangnam district in early October 2026, followed by AI remakes that cast other people and fictional characters in the three roles. This page lets you cast the roles with your own friends.
Who is Photo 1, Photo 2 and Photo 3
The clip has three people, and each one gets a photo of their own. Left and right are the screen's left and right in the opening frame, not the person's own sides. All three appear in both halves of the video, so every photo matters.
- 1
Photo 1
Centre of the opening frame
Holds the selfie camera, sings straight into the lens and leads the final group hook.
A face-forward portrait with a natural smile. This person is close to the lens for most of the video.
- 2
Photo 2
Left of the opening frame
Waves at the start, then takes a solo with chest taps, swaying and raised arms.
Show your upper body: the solo is hand-on-chest, hand-on-hip and big arm gestures.
- 3
Photo 3
Right of the opening frame
Waves with both hands, then spins in, tosses the hair and points at the camera.
Show your hairstyle clearly: the entrance is a spin, a hair toss and shoulder sways.
What happens, second by second
The original is one continuous shot with six beats. Knowing them tells you where each photo does its work: Photo 1 carries the opening, the close-up and the finale, so spend your clearest portrait there.
| Time | Shot | Who is on screen | What happens |
|---|---|---|---|
| 0:00 – 0:01.4 | The wave | All three | Photo 1 smiles into the selfie camera from the centre. Photo 2 waves and cheers on the left, and Photo 3 waves with both hands on the right. |
| 0:01.4 – 0:04.5 | The solo to the lens | Photo 1 | Photo 1 gasps, leans in, and the other two slip out of frame. What follows is a close-up of wide-mouthed singing, flashing teeth and a round O mouth. |
| 0:04.5 – 0:08.2 | The chest-and-hip solo | Photo 2 | The camera swings to Photo 2, who puts one hand on the chest and one on the hip, sways, thrusts a hand at the lens, then raises a fist and an open palm. |
| 0:08.2 – 0:11.7 | The spin and the point | Photo 3 | Photo 3 spins into frame, tosses the hair, mouths the words with swaying shoulders, points at the camera, waves and tosses the hair again. |
| 0:11.7 – 0:16.4 | The return | Photo 1 | Photo 1 steps back in from the left, reaches a hand toward the lens, tilts the head with a smile and leans in again for the build to the hook. |
| 0:16.4 – 0:19.1 | The group shout | All three | All three open their mouths for the hook, then end on a wide grin with hands in the air: Photo 1 in the centre, Photo 2 with a hand behind the head, Photo 3 waving. |
Photos that work, and photos that do not
Use
- ✓One adult, face fully visible and front-facing
- ✓Even daylight or soft front light
- ✓Head and shoulders, with a collar or shirt in frame
- ✓Neutral or lightly smiling expression
Skip
- ✕Sunglasses, masks or anything over the eyes
- ✕Group photos: the model has to guess who you mean
- ✕Heavy filters, beauty effects or tiny, blurry faces
- ✕Minors, celebrities or anyone who has not agreed
Choosing photos for each person
The model copies the face, hairstyle and clothes from your photos, so the pictures you pick matter more than anything you could type. For Photo 1 choose a portrait taken at arm's length, looking at the lens, with a natural smile: that person is closest to the camera for most of the video and ends it on a big grin. For Photo 2 choose a picture that shows the upper body, because the solo is all hand-on-chest, hand-on-hip and arm gestures. For Photo 3 choose one where the hair is clearly visible, since the entrance is a spin and a hair toss.
Photos that fail are predictable. Sunglasses and masks hide what the model needs to copy. Group pictures make it guess who you mean. Heavy filters and tiny faces pull the result toward an average face. Matching the angle across the three photos helps the fast swings between people look natural, and so does similar light. If the first result is wrong, change the photo rather than the settings. Try the standard version first and spend the HD price on the cast you keep.
Want to know who is behind the site? Read the About page. Wondering what a credit buys? See pricing.
When a result looks off, and what to change
The cuts and the camera swings are copied from the original clip, which is what makes the video land on the beat and also what makes it sensitive to the photos. If one moment looks wrong, change the photo for that moment rather than anything else.
- The switch from Photo 1 to Photo 2 looks awkward
- The camera swings fast there. Use a Photo 2 portrait taken at a face-forward angle like Photo 1, with similar light.
- Photo 3's spinning entrance is hard to follow
- Choose a picture where the face and the hairstyle are both clear. The spin and hair toss need something to copy.
- Photo 1's open-mouth expressions are not what you expected
- The close-up has a surprised O mouth and a wide-mouthed song. A portrait looking straight at the lens gives the model the most to work with.
- The model kept someone from the original clip
- This happens when a photo is not a single clear person. Use one person per photo, with the face fully visible and nothing covering it, and try again.
Who makes a Gangnam Style AI video, and why
Three friends, one night out
The format is made for a trio: the person who always holds the phone, the one who dances and the one who flips their hair. Drop in the real three and send the clip to the group chat before the evening is over.
A birthday or a send-off
Cast the guest of honour in the centre and two people who love them on either side. A 19-second video with a song everyone knows is easy to share and easy to watch twice.
A team or a club intro
Three colleagues, three teammates, three founders. The solo turns give each person a few seconds where the camera is on them alone, which is why the format works as a light-hearted team introduction.
A creator testing a trend
If you post trends, the question is how to be different from the thousand other versions. A cast nobody has seen is the answer, and a retry costs a standard render, not an afternoon of filming.
Gangnam Style AI compared with a generic face-swap template
Many Gangnam Style tools put one face on one dancer. That works for a solo joke, but the trend is a trio with a passing spotlight. Here is what is different on this site.
| Gangnam Style AI (this site) | A generic face-swap template | |
|---|---|---|
| People you can cast | Three, each with their own photo | Usually one face on one dancer |
| What you write | Nothing. The prompt is built in and tuned to the clip | Often a prompt of your own, or none |
| The camera moves | Copied from the original: swings, spin-ins, group finish | Depends on the base clip |
| Sound | The original sound is laid under the finished video | Varies; sometimes silent |
| Price before sign-in | Shown on the page: from $9.99 | Often only after you register |
| If it fails | Credits are returned automatically | Varies |
Prices you can see before you sign in
Credits are bought once and never expire. A standard video (480p) is 100 credits, an HD video (720p) is 300. If a video fails, the credits are returned automatically.
Single
$9.99
1 standard video
Trio
$24.99
3 standard videos, or 1 HD video
Full band
$59.99
8 standard videos, or 2 HD and 2 standard
Use it responsibly
- Use only photos of yourself or of adults who have agreed to appear. You confirm this before every video.
- Do not upload minors, celebrities or public figures, and do not use a video to mock, harass or impersonate anyone.
- Say that the video is AI-generated when you post it.
- Your photos are used to make your video and are not kept with it beyond the run, unless you agree otherwise. See the Privacy page.