I recently built a small web tool for a trend that has been all over my feed. Near the end of Drake's "Burning Bridges" music video there's a party in a chandelier-lit dining room: he holds a flat hand up by his forehead and looks around the room (people call it the "where she at" move), then raises a silver cup and waves at the camera. Since late September people have been using AI to recast that party scene with themselves, their friends, their pets and even TV casts like The Office. I followed the usual tutorials to make my own and kept fighting with clip cutting, written instructions and retries, so I built AI Burning Bridges Video to make it a one-photo job. Sharing it here in case anyone wants to join the trend, and I'd really appreciate feedback.
About AI Burning Bridges Video
AI Burning Bridges Video turns one photo into the Burning Bridges party scene. It works on a recreated version of the scene: the same chandelier-lit room, the same handheld camera moves and the same choreography, with stand-in party guests behind the lead. The AI edits that clip so the lead becomes the person in your photo from the first frame to the last, while the room, the guests and the camera stay exactly as they are. The finished video plays with the song from that part of the music video. If you missed the meme, the Burning Bridges trend page has the backstory and how it spread.
Features
- One photo, nothing else: no prompt to write, no clip to cut, no editing timeline.
- The whole person, not a face sticker: face, hair, outfit and even hands come from your photo.
- Same room, same moves, same sound: reaching out to the crew, the hand over the eyes, the cup-and-wave finish, with the song underneath.
- Three lengths: 5 seconds (just the "where she at" moment), 15 seconds (the best-known stretch) or 28 seconds (the full party scene), in Standard 480p or HD 720p. The 5-second clip is Standard only.
- Widescreen 16:9 MP4: the same frame as the music video, and it posts fine to TikTok, Reels, Shorts and X.
- Ready in minutes: about 3 minutes for 5 seconds, 5 to 10 minutes for 15 or 28 seconds.
- Works in the browser: iPhone, Android or desktop, with no app to install.
- Failed renders are never charged: if a render fails or a photo is filtered, the credits go straight back to your balance.
- Honest pricing: there's no free trial because every video uses real GPU time. A one-time $4.99 pack covers one 5-second clip, and plans start at $9.99/month for one 15-second Standard video. Four sample videos (a 1920s gentleman, a party queen, a Texas cowboy and a golden retriever in a tuxedo) show what you get before paying.
Doing it by hand vs. with a tool
If you'd rather build it yourself, the popular route is: cut the party scene from the video, run it through a general AI video tool such as Higgsfield (its Genjutsu character swap), Kling or Dreamina with your photo and a written instruction, then add the sound back in CapCut or TikTok. That gives you more creative control and works for other scenes too, but tutorials warn it often takes three to five tries before the room, the crew and the moves stay put. I wrote up the honest differences in Burning Bridges Higgsfield vs. one-photo tool, so pick whichever fits how much time you want to spend.
How to use?
Website: https://aiburningbridgesvideo.com
- Open the site and upload one photo. One person, face clearly visible, good light. A half-body photo with both hands showing works best, because the AI copies your hands too.
- Choose 5, 15 or 28 seconds, then Standard or HD. You see the cost before you generate.
- Click Generate, wait a few minutes, then download the MP4. Many people post it with the official "Burning Bridges" sound from the TikTok or Instagram library.
There's also a step-by-step guide, How to Make a Drake AI Video, with a photo checklist and posting tips. Birthday clips for dads and friends have been the most fun so far.
It's an independent fan project, not affiliated with Drake or his label. Please only use photos of yourself or of people who agreed; photos of celebrities get rejected. Feedback, bug reports and ideas are all welcome, thanks for taking a look!