The around the corner trend is a 12-second video in which friends hold back an angry man while a calm person sips a drink and the real fighter walks out from around the corner, and you can make it from one photo on the around the corner page. This guide answers the questions that come after the basics: who the five people are, which photos work, what it costs and how to post it. videosi is independent and has no connection with TikTok, CapCut or any video you may have seen.
Key takeaways
- The format is built on the line that you are always calmer when you know who is around the corner.
- The earliest clips we found are from late August 2026, so it is spreading, not brand new.
- You need one photo, the fighter, and can add two more for the calm one and the hothead.
- The 12-second MP4 comes with the trend's original sound, and it is played as slapstick with no weapons.
- A 12-second video on the default model costs 10 credits, and failed renders are refunded.
In this guide
- What is the around the corner trend, and where did it start?
- What are the five shots of the clip?
- Who are the five people, and which ones can be me?
- Which photos work best?
- Does the clip come with the original sound?
- Is the fight violent?
- How much does it cost and how long does it take?
- Why does one of the people look wrong or missing?
- Can I use a celebrity photo, a friend or a pet?
- How is this different from doing it in ChatGPT, Sora or Kling?
- Is the trend still going, and should I post now?
What is the around the corner trend, and where did it start?
The around the corner trend is a short comedy format built on one idea: you are always calmer when you know who is waiting around the corner. Two friends drag an angry man back, a calm person sips a drink without a care, and then the real fighter steps out from around the corner.
The earliest clips we found on TikTok date from late August 2026, when a school-corridor version appeared with the caption "why is bro so calm". AI versions with football stars and fighters followed in September 2026. We could not identify a single original post, so treat any claim about the first creator as unconfirmed.
What are the five shots of the clip?
Almost every version follows the same order. First a wide shot of the scuffle, with the angry man lunging and two friends grabbing him. Then a close shot of the arm-hold while he shouts and points. Third, a cut to the calm person leaning on a wall with a small cup.
Fourth, the calm person keeps sipping in the foreground while a small figure steps out from around the corner behind. Last, that figure walks toward the camera, tense, flexing both fists. The whole sequence runs about 12 seconds, which is the length of the clip you get on the around the corner page.
Who are the five people, and which ones can be me?
The clip has five people: the hothead, two friends who hold him, the calm one and the fighter who walks out. On videosi, Photo 1 is the fighter, the person who steps out from around the corner, and it is the only photo you must add.
Photo 2 is the calm one with the drink and Photo 3 is the hothead being held back, and both are optional. Leave them empty and a generated adult takes the role. The two friends who do the holding are always generated, so a clip never has more than three faces from your own photos.
Which photos work best?
Pick sharp, front-facing photos with the head and shoulders in frame. The walk-out ends close to the camera, so the fighter's face is on screen longest and carries the whole joke.
If you add three photos, choose people who look different from each other, with different hair or clothes. The roles are told apart by the photos, and two similar faces are the most common reason they blend. Use adults who agreed to appear, never children.
Does the clip come with the original sound?
Yes. The video model makes the picture, and videosi then lays the trend's original sound on top, so the downloaded MP4 already has it. You can post it as it is or replace the audio with the exact sound of a clip you saw inside TikTok, Reels or Shorts.
There is no single named song. The clips we checked share one reused original sound, which footballer, fighter, idol and horror versions all use, and we have not confirmed a title or an artist. We do not guess one.
Is the fight violent?
It is played as slapstick. There is no blood, no injury and no weapon, and the scuffle in the first seconds is two friends holding someone back, not a brawl. The fighter at the end only walks and flexes.
The prompt also asks for five adults in total and no text or logos. Because it shows real faces in a staged confrontation, add the platform's AI label when you post it.
How much does it cost and how long does it take?
A 12-second clip on the default Standard model at 720p is 10 credits, and an account's first video is capped at 10 credits. A Single pack of 10 credits is $6.99, and there is no free tier. Higher tiers and resolutions cost more, and the price is shown before you press Generate.
A render usually takes a few minutes. If it fails, times out or is blocked, the credits come back automatically.
Why does one of the people look wrong or missing?
Five people in one scene is the hardest part for any video tool. The usual faults are blended faces when two photos look alike, a friend who looks like one of your photos, and hands that bend in the tangle of arms during the scuffle.
Try clearer photos, change which photo is Photo 1, or leave the optional photos empty so there are fewer faces to keep apart. Likeness is never guaranteed, and each attempt uses credits.
Can I use a celebrity photo, a friend or a pet?
Friends and co-workers who agreed are the intended use: the trend is an inside joke between people who know each other. Many popular versions on TikTok use footballers and fighters, but a clip made from someone else's face may be removed by a platform, so only use photos you have the right to use.
The page is built for people. Pets and cartoon characters are not part of this effect, and videosi does not accept photos of children.
How is this different from doing it in ChatGPT, Sora or Kling?
You can try those tools, but expect several attempts. General video tools rarely keep five people apart across a scuffle, a calm cut and a walk-out, so faces blend or someone disappears. They also do not know the shot order that makes the joke work.
videosi builds the whole sequence from one to three photos in one go, using a reference of the original shots, and puts the trend's sound on the result. You write no prompt.
Is the trend still going, and should I post now?
As of November 2026 it is spreading: the first clips are from late August and September, and new versions keep appearing, some with tens of thousands of views. Format trends like this one usually move to new casts quickly, so the easiest way to stand out is to use people your own audience knows.
Post within a few days of making the clip, keep the original sound and add the AI label. Remember that views and dates on this page come from clips we checked ourselves and are not official figures.
Make your own
Open the around the corner AI page, add the photo of whoever steps out from around the corner, and press Generate. The result is a 12-second clip you can post straight away.
