# How to Make the Burning Bridges AI Video: The Drake Party Trend, Second by Second — AI Rap Duo

Canonical page: https://airapduo.io/guides/how-to-make-the-burning-bridges-ai-video
Published: 2026-10-10
Last updated: 2026-10-10

> How to make the Burning Bridges AI video with yourself at the chandelier party, where the trend came from, what happens in each second, and what our own test renders taught us.

The Burning Bridges AI video is the clip of someone at a crowded party under crystal chandeliers, a silver cup in one hand, pointing at friends, shading their eyes to look around the room on "where she at", then waving at the camera. The scene comes from the end of Drake's "Burning Bridges" video, and since September people have been putting themselves, their friends and TV characters into it with AI. Below is where the trend came from, what each second of it does, and three ways to make your own.

(Image: The photo behind one of our examples (an AI-generated face) and a frame from the finished video.)

## Where the Burning Bridges AI trend came from

"Burning Bridges" is track eight on Drake's album *Iceman*, released on 15 May 2026. Its video ends at a party in the private dining room of a restaurant: a low ceiling, chandeliers hanging just above the guests' heads, green light along the walls and a room packed shoulder to shoulder. Drake holds a silver cup, points at his friends, covers his eyes with one hand as if looking for someone on the line "where she at", and waves.

On 19 September 2026 the first AI version appeared on Threads, made with a character-swap tool. Within days TikTok, Instagram Reels and YouTube Shorts filled up with remixes: Michael Scott, Tony Soprano, Walter White, Thanos and plenty of ordinary people at the same party. By 10 October a single Instagram compilation had passed 100,000 likes, and the official video was getting a third more daily views than at the end of September, mostly from people who had seen the remixes first. This site isn't affiliated with Drake, his labels or anyone involved in the song.

## What happens in a Burning Bridges AI video, second by second

We went through the original party scene and the most-shared remixes frame by frame. Every good version keeps the same fifteen seconds, in one handheld take with no cuts:

1. **0–3 s: vibing with the cup.** The cup comes up toward the lens, then a nod to the beat and a hand to the collar.
2. **3–8 s: hyping the friends.** A turn to the crowd, a point, a reach; the friends cheer back.
3. **8–10 s: the cup to the lens.** A turn back and a hand pushed right at the camera.
4. **11–13 s: "where she at".** A hand flat over the eyes like a sun visor, scanning the room, then a wave.
5. **13–15 s: still partying.** The camera eases back and the crowd keeps going.

(Image: Four frames from one vertical Burning Bridges AI video in a dim room with crystal chandeliers: a woman in a green top holding a silver cup and touching her chest, reaching out to the cheering crowd, pushing her hand at the camera, and holding a hand over her eyes — Four moments from our single-photo vertical example.)

The song matters as much as the moves. Most clips use the part of the track where "damn, where she at" lands on the hand over the eyes, so the gesture and the line hit together.

## 3 ways to make your own Burning Bridges AI video

**1. A face swap on the original footage.** Character-swap tools put your face on the real clip. It looks exactly like the video, but you often end up in the original outfit, the result is usually widescreen and low resolution, and you are working with someone else's footage.

**2. A party template.** Several apps use one AI-made party clip and swap each user in. It is quick and often vertical, but the room tends to be a bright golden ballroom that looks nothing like the original, and every video has the same crowd.

**3. One photo on our page.** On [Burning Bridges AI](https://airapduo.io/burning-bridges-ai.md) you add one photo and pick vertical or widescreen. We copy the moves and the camera from the trend second by second, build the room fresh around you, grade it down to the original's mood and add the song.

## How we copy the moves without copying the footage

Our version never uses a frame of the original. We turned the fifteen seconds into a depth map: a moving picture of how far each part of the scene is from the camera, with shapes and motion but no faces, colours or clothes. The video model reads it as choreography and camera movement and paints a new scene on top, with you in the middle.

(Image: Two pairs of frames: a coloured depth-map silhouette with a hand over its eyes and a cup in the other hand, next to a woman in a dim chandelier room doing the same with a silver cup — The depth map (left of each pair) only carries the moves; the room, the crowd and the person are new.)

That is also why your video keeps your own face, hair and the clothes from your photo, and why the cup stays in your hand: the model draws you holding it from the first frame instead of pasting a face onto somebody else's arm.

## What our Burning Bridges AI video tests got wrong

We ran our test renders at 480p before making any example, and compared every one with the original frames side by side.

- **A plain text prompt didn't feel like the video.** Our first attempt described the scene in words only. The moves were close, but it looked like any AI party; without the original's choreography the joke was gone.
- **Swapping a person in lost the cup.** Putting a new person into a ready-made clip removed whatever they were holding, so we generate each video around the person instead.
- **Every render came out too bright.** The model kept lighting the walls however dark we asked it to be. A colour grade after the render brings it back to chandeliers and green light, with the face still clear.
- **Pets grew hands.** A dog standing in for a person waved with human fingers. Until that is fixed, our Burning Bridges AI video takes one photo of one person.
- **Caps with logos.** The crowd sometimes wears caps with letters on them however plainly we ask; it's rare and small, and we are still working on it.

Our test renders took between 3 min 6 s and 4 min 50 s, plus about half a minute for the grade and the song.

(Annotated frame: A frame from our widescreen example, which matches the shape of the original video. Notes: You, in the clothes from your photo, cup in hand; A packed, cheering crowd; Crystal chandeliers just above head height; Green light along the walls, like the original.)

- Example video: Burning Bridges AI: The Party, Vertical. From one photo: he touches his collar, laughs with the friends around him, looks for her with a hand over his eyes on the line and raises his cup, holding it the whole time. Video: https://media.airapduo.io/examples/bridges-man.mp4
- Example video: Burning Bridges AI in Widescreen. From one photo, in widescreen like the original: the packed chandelier room, the cup up, the point at the crowd, the hand over the eyes on "where she at" and the wave, with the song. Video: https://media.airapduo.io/examples/bridges-wide.mp4
  Two of our examples: a vertical one for TikTok and a widescreen one like the original. Tap to play the 480p cut with sound.

**[Make your Burning Bridges AI video](https://airapduo.io/burning-bridges-ai.md)** One photo, every move of the trend in time with the song, vertical or widescreen, ready in about 5 minutes.

## Tips for the photo you upload

- **One person, face clearly visible.** A front-facing photo in daylight keeps the likeness best. No sunglasses, no masks, nobody else in the shot.
- **Waist-up is ideal.** The video is framed from about the waist up, so a waist-up photo gives the model your clothes to work with.
- **Adults only.** Photos of children are turned away.
- **Pick the shape before you start.** Vertical 9:16 fills a phone for TikTok and Reels; widescreen 16:9 matches the original video and looks right on YouTube.

## What a Burning Bridges AI video costs

Face-swap tools and templates usually run on credits or a subscription. Ours is paid per video: the 15-second Burning Bridges AI video at 720p is 20 credits, $7.99 once or about $6 out of the $14.99 pack of 50 credits. No subscription, and a failed render refunds itself. See the [pricing page](https://airapduo.io/pricing.md).

| | Face swap on the original | Party template | Our Burning Bridges AI |
|---|---|---|---|
| Your clothes | Often the original outfit | Often the template's | Yours |
| Shape | Usually widescreen | Usually vertical | Vertical or widescreen |
| Quality | Often 480p | Often 480p | 720p |
| The song | Usually included | Usually included | Included |
| Cost | Credits or a subscription | Credits or a subscription | $7.99, or about $6 from a pack |

## Burning Bridges AI video FAQ

### What song is in the Burning Bridges AI video?

Drake's "Burning Bridges", from the album *Iceman* (2026). The clip uses the part where "damn, where she at" lands on the hand over the eyes.

### Can I put myself in the Burning Bridges video?

Yes: swap your face onto the original clip, use a party template, or make a new version with your own face and clothes from one photo on our [Burning Bridges AI](https://airapduo.io/burning-bridges-ai.md) page.

### Is my video the original footage with my face on it?

Not with ours. We copy only the moves and the camera; the room, the crowd and you are generated fresh.

### Can I make it with my dog?

Not yet. The moves are a person's, and in our tests a dog in that role ended up with human hands, so for now it's one photo of one person.

### Can I post it on TikTok?

Yes. The song is already in the video. You can also pick the trending sound in TikTok and turn ours down; the moves are in time either way.

**[You're on the guest list](https://airapduo.io/burning-bridges-ai.md)** Your photo, the chandelier party, the cup, the look and the wave in 15 seconds. One video $7.99, no subscription.

## More
- All guides: https://airapduo.io/guides.md
- All pages: https://airapduo.io/index.md
- Pricing: https://airapduo.io/pricing.md