The BABYMONSTER Deserve AI trend is a short clip in which you sing the DESERVE chorus in a recording booth, made from one photo of you. It is not footage of BABYMONSTER and it does not turn you into a member. On videosi you get a 12-second video with the trend's original sound, with no prompt to write. This guide goes past the basics into the song's background, the booth, the gestures, the photo and the limits.
Key takeaways
- DESERVE is a slow pop ballad released as a digital single on 23 October 2026, so the trend is about feeling, not choreography.
- You only get the recording booth version; the street and home clips on the page are examples, not options.
- The clip has the original sound but only rough mouth sync, and the voice is not yours.
- A default 12-second video costs 10 credits, and a failed render returns them.
- Use a bright photo of an adult, and label the post as AI.
In this guide
- Where did the Deserve challenge come from?
- Why does a ballad work as a video trend?
- Why a recording booth, and can I pick another place?
- What is happening with the hands, and why does it matter?
- Which photo gives the best result in a dark side-on shot?
- Does the clip keep anything of the original singer?
- What does the sound on the finished clip actually do?
- How many credits does one clip take, and what if it fails?
- How should I post it, and do I need to say it is AI?
Where did the Deserve challenge come from?
It grew out of a slow song rather than a dance. DESERVE was released as a digital single on 23 October 2026, at 1 PM KST according to KPopOfficial, which lists only the title track. YG Entertainment announced it ahead of time as a slow-tempo comeback, as The Korea Times reported on 14 October, and it arrived during the group's world tour.
Within days, people were filming themselves singing the chorus, and the earliest tagged posts we found are dated 26 October. We could not identify one original post, so we do not credit anyone as the creator. The trend is still young, which is why the page calls it emerging.
Why does a ballad work as a video trend?
Because the song asks for a performance, not a skill. Bandwagon describes DESERVE as a pop ballad with lyrical synth melodies and a retro mood, bittersweet and restrained until an emotional climax. It also says the group focused on emotional flow rather than vocal range or high notes. That is easy to copy: you close your eyes, lift your chin and let the feeling show.
It also explains why the trend splits into two moods. Some people sing it sincerely into a microphone, and others turn it into comedy by screaming the high notes outdoors. Both fit a 12-second clip, because the chorus does the work and you need no setup.
Why a recording booth, and can I pick another place?
The booth is the version we built, and it is the only one you can get. The finished clip is a dark recording booth with a hanging studio microphone, headphones on, a close opening shot that eases back to chest height, and a side-on camera. The street and home clips shown on the page are examples of how other people play the trend, not scenes you can choose.
The reason is practical. A fixed place keeps the shots and gestures stable, which is what makes the clip read as a heartfelt take instead of random movement. If you want a different setting, the BABYMONSTER Deserve AI page is not the right tool, and we would rather say so than promise it.
What is happening with the hands, and why does it matter?
The gestures carry the emotion, so they are the part we protect most. Across 12 seconds, the open hands rise into frame, one presses to the chest while the other stays open below it, and then both turn slowly in front of the body while the long notes are held. The breakdown on the BABYMONSTER Deserve AI page lists the beats from 0 to 12 seconds.
Without a motion reference, a video model usually turns slow gestures into random waving. We supply a reference clip for the motion and your photo for the person, so the hands follow a deliberate path. If your photo shows your arms, the sleeves and skin tone also hold up better when the hands come up.
Which photo gives the best result in a dark side-on shot?
A bright, sharp one with your whole face visible. The booth is dark and the camera is side-on, which is where a face drifts most, so a dim or blurred selfie gives the model too little to hold. Skip sunglasses, masks and hair across the face.
- Pick a photo where you face the camera, with even light.
- Include your shoulders and, if you can, your arms.
- Wear what you want to see, since your top is on show for most of the clip. Plain tops and simple prints come out cleanest.
Use a photo of an adult, and only of yourself or someone who agreed. The clip is built for one person.
Does the clip keep anything of the original singer?
It should not, and the prompt is written to prevent it. The reference video shows a young man in a printed raglan T-shirt, and the instruction is to replace him completely, so his hair, shirt and sleeves do not carry over unless your own photo shows the same. This matters because video models treat a reference loosely by default, and leftovers from the reference are a common complaint.
One more rule keeps faces honest: your face shape stays as it is in your photo. The model is told never to slim or narrow it toward the person in the reference. We have not test rendered this page yet, so likeness is not guaranteed, and a bad result returns your credits.
What does the sound on the finished clip actually do?
The MP4 carries the trend's original sound, so it is not silent, and there is nothing to line up before you post. Two limits are worth knowing. Your mouth follows the words only roughly, so it is not exact lip sync. And the voice is the song's, not yours.
That is why the clip works best as a mood piece: eyes closed, hands moving, the chorus underneath. If you want to be seen singing a specific line in sync, this is the wrong format, and filming it yourself will beat any AI clip.
How many credits does one clip take, and what if it fails?
Plan on 10 credits for the default. A 12-second video at Standard 720p costs 10 credits, so the $6.99 Single pack with 10 credits covers one. The other tiers cost more at 720p: 17 for Fast, 20 for Pro and 37 for Ultra. Credits never expire, and there is no free generation.
A failed, timed-out or blocked render returns its credits automatically. Since the page has not been test rendered, we suggest starting on the default tier with a good photo before paying more for a higher one. A better tier does not fix a weak photo.
How should I post it, and do I need to say it is AI?
Yes, switch the AI label on when you upload to TikTok or Reels. The clip is a synthetic video of you, and the label keeps it honest. Choose 9:16 for TikTok and Reels, or 1:1 or 16:9 for other places, before you generate.
Say plainly that it is a fan trend clip. videosi is independent and not affiliated with BABYMONSTER, TREASURE or YG Entertainment, and it never recreates a group member. Only post it if you have the right to use the photo.
Ready to try it? Open the BABYMONSTER Deserve AI page, upload one bright photo and generate your booth clip.
