Quick answer
Upload two photos to Starrd's White Chicks template and it recreates the car scene from the 2004 comedy White Chicks with the two of you in it. Photo 1 is the passenger on the left, holding a small Pomeranian and giving a horrified side-eye. Photo 2 is the driver on the right, the one joyfully singing Vanessa Carlton's "A Thousand Miles" at night. It plays as a wide two-shot through the windshield, a close-up of the passenger's side-eye, a close-up of the driver belting "and I need you… and I miss you" ending with a finger tapped to the temple, then the passenger's frozen stare. Works for any two people, the Pomeranian is generated, and nobody's appearance is changed. 18 seconds, horizontal 16:9, 250 credits (about $12.50).
The same dance, any subject — tap one to make it with your photo.
Two photos in, and out comes the car scene from White Chicks: one of you behind the wheel at night belting Vanessa Carlton's "A Thousand Miles", the other in the passenger seat clutching a Pomeranian and staring in horror. Here's exactly what you get:
White Chicks
Two photos. One of you sings A Thousand Miles at the wheel, the other holds the Pomeranian and stares. 18 seconds, horizontal.
What You Get
- The wide shot. A two-shot through the windshield at night: the driver on the right, the passenger on the left with a small Pomeranian in their arms.
- The side-eye. A close-up of the passenger, horrified, eyes sliding toward the driver.
- The big note. A close-up of the driver belting "and I need you… and I miss you", ending with a finger tapped to the temple.
- The frozen stare. Back to the passenger, who hasn't moved.
- Any two people. A friend, a partner, a sibling, a parent. Nobody's appearance changes: "White Chicks" is only the name of the movie.
- 18 seconds, horizontal 16:9, the same widescreen framing as the film.
What the Trend Actually Is
The scene is from the 2004 comedy White Chicks. At night, the driver, Latrell (played by Terry Crews), joyfully sings "A Thousand Miles" by Vanessa Carlton ("Making my way downtown…") to a front-seat passenger who is clutching a small Pomeranian and clearly wishes they were anywhere else.
It spread on TikTok through a repost by @reelscenes10, at around 50K plays, captioned "One of the best scene😂" with #terrycrewswhitechicks and #thousandmiles.
Why it works: the contrast. One person is having the time of their life; the other is trapped in a car with it. Everyone has been on one side of that.
Why it suits AI: you already know who in your life is the driver and who is the passenger. Two photos and you cast it.
Which Photo Goes Where
The photo order is the seating plan:
- Photo 1: the passenger (left). Holds the dog, gives the horrified side-eye, freezes.
- Photo 2: the driver (right). The one singing.
Put the person who would actually sing in the car in slot 2, and the one who would suffer through it in slot 1. The Pomeranian is generated; you don't need a dog.
Use clear, well-lit, front-facing photos, one person per photo. Head and shoulders is ideal. A second face in either photo makes it unclear who goes in that seat.
How the Video Plays Out
The template runs the scene in one 18-second horizontal clip:
- Wide two-shot through the windshield. Night, both of you in the front seats, the dog in the passenger's arms.
- Close-up: the passenger's side-eye.
- Close-up: the driver belting "and I need you… and I miss you", ending with a finger tapped to the temple.
- The passenger's frozen stare.
The Fastest Way
White Chicks
Two photos, the full 18-second car scene. The driver sings, the passenger suffers. No prompt to write.
Two photos, 250 credits (about $12.50), no prompt. The template holds the car, the night, the dog, the camera and the performance; your photos only decide who's in which seat.
Build It Yourself
If you'd rather wire it up by hand, the shape is a two-person scene with fixed seats: a still that sets up the car and who sits where, then a video pass that performs the song.
1. Make the still first. One horizontal image of the two people in the front seats of a car at night, seen through the windshield. This is where the casting is decided (who drives, who holds the dog), and it's cheap to iterate on. Generate it on its own and look at it before you pay for any video.
A cinematic night-time shot through the windshield of a car, two people in the front seats. On the LEFT, in the passenger seat: the person from reference photo 1, holding a small fluffy Pomeranian in their arms, expression horrified, eyes sliding sideways toward the driver. On the RIGHT, behind the wheel: the person from reference photo 2, mid-song, joyful, fully committed. PRESERVE both people's exact identities, faces, hair and their own clothing; do not change either person's appearance. Warm streetlight passing across their faces. Horizontal 16:9. No text, no logos, no watermarks.
2. Then animate it with a reference-to-video model. Feed the still in with the two identity photos and a time-segmented prompt for the four beats: wide shot, passenger side-eye, driver close-up singing with the finger tap to the temple, passenger frozen. Seedance 2.0 is the safest choice for keeping two faces consistent through cuts.
Check the seats before you commit. If the model gives the song to the passenger, or the dog to the driver, fix the still, not the video prompt.
Common Mistakes
Swapping the slots. Photo 1 is the passenger, photo 2 is the driver. Put your singer in slot 2.
Group photos. One person per photo. Two faces in one slot confuses the casting.
Expecting a makeover. The template doesn't change how anyone looks. "White Chicks" is just where the scene comes from.
Cropping it vertical. It's built horizontal, like the film. Cropping to 9:16 cuts someone out of the two-shot.
Posting it muted. The song is the whole joke. Sound on.
FAQ
How many photos? Two, one per seat.
Which is which? Photo 1 is the passenger (left, holds the dog). Photo 2 is the driver (right, sings).
Do I need a dog? No. The Pomeranian is generated.
Does it change how we look? No. Nobody's appearance changes.
How long is it? 18 seconds, horizontal 16:9.
What does it cost? 250 credits, about $12.50.
Related Reading
- How to Make the Walk It Like I Talk It AI Video: another car singalong, with four of you in the seats
- How to Make the On The Radar AI Video: two photos, two performers
- How to Make the Tonight We're Getting Naughty AI Video: a two-person routine from two photos
- How to Make the Storm Dance AI Video: the newest one-photo trend
- Viral AI Video Trends of 2026: what's climbing right now