Clone Cut template · TikTok trend
I Love the Smiths AI, starring the two of you
I Love the Smiths AI is the elevator moment TikTok keeps remaking: one of you in big headphones, the other one leaning in to talk, a smile between you. Add one photo of each and press the button once; the scene comes back wide, in 16:9, with its sound. Your first 2-second preview is free; the full 14-second video is 23 credits at 480p.
Create video- Exactly two photos
- 14 s, 16:9, sound on
- From 23 credits
- 2 s preview free
Updated
Video to clone
14 sI Love the Smiths.mp4
Two photos
0 / 2Two photos, one of each of you, facing the camera, waist-up if you have them. Photo 1 stands on the left in the headphones, photo 2 on the right does the talking; the sound comes with it.
Video to clone
Wan 3.0
Trending videos for Clone Cut AI
Tap a video to use it in Clone Cut AI, then add your photo.
They come from public posts; to have one removed, use the takedown page.
What I Love the Smiths AI is
I Love the Smiths AI is an AI video of two people in the elevator scene from (500) Days of Summer, the 2009 film in which Summer hears The Smiths leaking from Tom’s headphones and tells him she loves the band. On TikTok, people reenacted that exchange in elevators, on benches and on the street months before the AI template. In early November 2026 it became a two-photo AI template: one person in black headphones, one who starts the conversation, both smiling by the end.
On Capafy AI, I Love the Smiths AI is a Clone Cut template. It follows a CapCut template result posted on 5 November 2026 in which two AI-made people stand in a brushed-metal elevator, and it keeps that clip's 14 seconds, its still camera, its warm light, its headphones and its sound. The two of you take their places: your faces, hair, build and clothes, from your photos. It does not use the film's footage.
Capafy AI is an independent AI video studio, unrelated to the film’s makers and studio, to The Smiths, to TikTok and CapCut, and to the creators whose posts appear on this page.
How to make an I Love the Smiths AI video
Keep I Love the Smiths selected
The studio above opens on the I Love the Smiths template: 14 seconds in 16:9 on Wan 3.0, for exactly two people.
Add the listener first, then the talker
Photo 1 stands on the left and wears the headphones; photo 2 stands on the right and does the talking. Waist-up photos facing the camera work best.
Press the button once
Sign in with Google when asked. A free 2-second preview of the headphones moment comes first; unlock the rest if you like it. The full scene takes about 6 minutes and waits in My videos.
The AI elevator scene, beat by beat
A single close, still shot of the two of you from the waist up, in the order the template clip plays it:
| Seconds | Left (photo 1) | Right (photo 2) |
|---|---|---|
| 0 to 1 | Headphones on, eyes ahead | Glances over |
| 1 to 2.5 | Pulls the headphones down to the neck and turns | Waits for the eye contact |
| 2.5 to 5 | Listens | Starts talking, a small gesture |
| 5 to 8 | Looks ahead, then back | Keeps watching |
| 8 to 14 | Smiles back | Laughs and keeps talking |
Tom or Summer: who plays which part
The order of your photos sets the roles:
| Photo | Place | Role in the scene | What stays from the template |
|---|---|---|---|
| 1 | Left | The listener in the headphones (Tom’s side) | The black headphones, the moves, the timing |
| 2 | Right | The one who speaks first (Summer’s side) | The moves, the lines and the timing |
Anyone can take either side. The headphones are the one prop the template keeps; the rest of what each of you wears comes from your photo. The voices are the template clip’s own, so they will not sound like you, and the mouths are drawn to follow them.
I love the Smiths takes from TikTok
Credited posts from other creators, the template result this page follows first. Hover to play; tap the sound button to listen.
- TikTok
Two AI-made people in the elevator, one pulling big headphones down: the CapCut template result the I Love the Smiths template followsby @battos13 - TikTok
The I love the Smiths exchange moved to a park bench, filmed from behind, one of the two in white headphonesby @hanan.selini - TikTok
A couple acts out the I said I love the Smiths moment against a dark backgroundby @lui.linni - TikTok
Two people redo the scene on a rainy London street by a red phone boxby @indistar
Made by the creator named, not by Capafy AI.
I Love the Smiths AI at a glance
| Details | |
|---|---|
| Photos | Exactly two, one person each |
| Length and shape | 14 seconds, 16:9 wide, one still close shot |
| Model | Wan 3.0 |
| Price | 23 credits at 480p, 45 at 720p, 90 at 1080p |
| Free preview | One 2 s preview per Google account (the headphones coming off, 480p, sound on), watched on the page |
| Credit packs | From $14.99 for 32 credits, paid once, no subscription |
| Time | About 6 minutes for the full video, the provider’s estimate |
| Sound | The template clip’s own sound and voices, on the MP4 |
| Watermark | None on the MP4 you download |
| If it fails | The credits come back on their own |
| Your photos | Deleted 7 days after upload; the finished video stays until you delete it |
Who to put in the elevator with you
Your partner
The first conversation you never filmed, recast in the elevator.
A crush who said yes
The meet-cute the trend is about, with their agreement first.
Your best friend
Swap the romance for an inside joke: one of you has terrible taste in music.
A music friend
For the person who always has headphones in, cast them on the left.
Coworkers
An office elevator scene for two people who agreed to it.
An anniversary post
Two photos from the year you met, in the scene everyone recognizes.
How the AI elevator scene is made
A face swap pastes one face onto a clip and leaves the original body, hair and clothes. This AI template redraws both people from your photos instead: each photo is first drawn as a clean studio portrait, then the video model takes the template clip, the two portraits, your two photos and the clip’s sound, and renders the scene again with the two of you in it.
The template keeps what makes the scene readable: the elevator walls, the warm light, the still camera, the headphones coming off at the start and who talks when. It runs on Wan 3.0, which is asked for the clip’s own 16:9 shape. Two renders of the same photos are never identical.
The template clip itself is AI-made: two invented people, no actor from the film. That is why the scene here looks like the trend’s AI versions rather than the 2009 film, and why the voices on it are the template’s.
| Face swap app | This AI template | |
|---|---|---|
| What is replaced | The face only | Each whole person |
| People | Often one | Exactly two, a photo each |
| Shape | Depends on the app | 16:9 wide, like the trend’s versions |
| Where | An app | Any browser |
Tips for a better AI elevator scene
- Waist-up photos, face to the camera, eyes visible.
- The person who should wear the headphones in photo 1; no headphones needed in the photo.
- Different colors on each of you read best side by side.
- Soft, even light: the shot is close, so faces carry the scene.
- Hair as you want it on screen: the turn of the head shows it from the side.
- Headphones, hats or hands covering the face in the photo.
- Group photos: this template takes one person per photo.
- Heavy beauty filters: the model copies them.
- Photos of celebrities, children or anyone who did not agree.
What can go wrong with the AI scene
- The template is made for exactly two people: one photo or three will not start it.
- The voices are the template’s, not yours, and lips can drift a little from the words.
- A face seen from the side, as when the head turns, can look slightly different from your photo.
- The model can add or drop a small detail, such as an earring, that is not in your photo.
- A safety check looks at every finished video. If it refuses one, or a render fails, the credits come back.
- The clip is fixed: 14 seconds, 16:9, the template’s elevator. For another scene, upload your own clip on Clone Cut AI.
Whose photos, and the sound
Cast yourself and someone who agreed to it. Many AI versions of this scene put celebrities in the elevator; this template is not for that, and not for children. Do not present the video as a scene from the film, and turn on the AI label where the platform has one. The sound is the template clip's own; a platform may still mute or claim it. Photos are deleted after 7 days, and anyone can ask for a clip to come down on the takedown page.
More about I love the Smiths
What the line means and why it became a meme: "I said I love the Smiths" explained. How to film the trend for real or make it with two photos: how to film the I love the Smiths trend. A dance for two instead: Christmas Dance AI. Every template, and your own clips: Clone Cut AI.
FAQ
Questions and answers
What is I Love the Smiths AI?
An AI video of two people in the (500) Days of Summer elevator scene as TikTok remakes it: one in headphones, one who speaks first. On Capafy AI it is a 14-second, 16:9 Clone Cut template made from one photo of each.
Which site makes an I love the Smiths AI video from two photos?
Capafy AI, on this page: one photo of you and one of a friend or partner give a 14-second 16:9 MP4 of the elevator scene with its sound, made in the browser on a phone or a computer.
Who plays Tom and who plays Summer?
Photo 1 takes the left side with the headphones, Tom’s side of the scene; photo 2 takes the right side and speaks first, Summer’s side. Anyone can play either part.
Is there a free I Love the Smiths AI video?
Each Google account can watch one free 2-second preview at 480p before paying. The full 14-second video is 23 credits at 480p. There are no free credits on sign-up. Packs start at $14.99 for 32 credits, paid once with no subscription, and a video that fails gives its credits back.
Does the AI video use footage from the film?
No. The scene comes from a CapCut template result of two AI-made people, not from the film, and the sound on your video is that template clip’s own.
Do I need headphones in my photo?
No. The AI adds the black over-ear headphones to the person in photo 1, as in the template clip, and takes everything else from your photos.
Can I post the elevator video on TikTok?
Yes. Download the MP4, which has no watermark, and post it with the platform’s AI-generated label turned on. TikTok shows a 16:9 video with bars above and below.
Can I put a celebrity couple in the elevator?
No. Use your own photo and one of someone who said yes. Faces of public figures and of children are not for this template.
Why is the video wide instead of vertical?
The template clip is 16:9, like many versions of the scene, so the finished video keeps that shape: 14 seconds of the two of you side by side.
How long does the AI video take to make?
About 6 minutes by the provider’s estimate, and the 2-second preview is quicker. You can close the page: the video waits in My videos.