Quick answer
Upload one photo to Starrd's Bad Times template and it turns you, or your pet, into the star of the viral heated-argument TikTok: bent over low, slowly rising, then hands on the head, pointing, 'hold on' hands, lunging into the lens and throwing the arms up, lip-synced to "BAD TIMES" by Imael Angel in one continuous handheld take. It keeps the background from your own photo, runs 30 seconds in vertical 9:16 on Seedance 2.5, and costs 250 credits (about $12.50). Pets are stood up on their hind legs but stay real animals: paws stay paws, birds keep wings and beaks, and the mouth is the animal's own.
One photo in, and out comes you (or your pet) having the whole over-the-top argument from the "BAD TIMES" TikTok, lip-synced to the song, beat for beat, in your own room. Here's exactly what you get:
Bad Times
One photo. You or your pet, losing it at the camera to Imael Angel's BAD TIMES, beat for beat, in your own room. 30 seconds, vertical.
What You Get
- The whole argument, beat for beat. Hands on the head, pointing, "hold on" hands, lunging right into the lens, arms thrown up.
- Lip sync to "BAD TIMES" by Imael Angel, with the song underneath the whole way.
- A slow-burn opening. It starts with the subject bent over low, head down, slowly rising to standing before the argument kicks off.
- One continuous handheld take. No cuts, like the original.
- Your own place. The background from your photo is kept and extended to fit, so it's your kitchen or living room, not a stock set.
- 30 seconds, vertical 9:16, ready for TikTok, Reels and Shorts. No text on screen.
What the Trend Actually Is
The source is a TikTok by @bouuhyachaka, past 10 million views, captioned "Ça va barder!!" A woman has a heated, completely over-the-top argument with whoever is holding the phone, lip-syncing to "BAD TIMES" by Imael Angel the whole time. She grabs her head, points, throws up "hold on" hands, lunges right into the lens and flings her arms up, all in one continuous handheld take. The original carries a "DISPUTE" label on screen; Starrd's version leaves the frame clean.
Why it works: it's a fight with the viewer. The camera is the other person in the argument, so every point and every lunge lands on whoever is watching. And the gestures are big enough to read on a phone at a glance.
Why it suits AI: the joke scales with who's doing it. A person losing it at the camera is funny. Your cat doing the exact same routine, pointing a paw at you like you ate its dinner, is funnier.
Why This Is Motion Copy, Not a Prompt
You can't describe your way to this video. Ask a model for "someone arguing with the camera" and you get an argument: generic gestures, a different one every render, and nothing timed to the song. The trend is this exact sequence, landing on this exact song.
So the template copies the motion instead of inventing it. It takes two things:
- The original clip, which supplies every gesture, the handheld camera and the timing.
- Your photo, which supplies who is doing it and where.
The performance is mapped onto your photo beat for beat, so the head grab, the point and the lunge hit where they hit in the original, and the mouth follows the lyrics.
Pets Stay Real Animals
This is the part most AI pet videos get wrong, and the part this template is built around. When you hand a pet to a human performance, the easy failure is a cat with human hands, or a dog with a person's mouth pasted over its face. Bad Times doesn't do that:
- Stood up, still an animal. The template stands your pet up on its hind legs so it has a body to argue with, but it stays your pet.
- Paws stay paws. Where a person would point or spread their fingers, the pet makes a whole-paw gesture. Never fingers, never human hands.
- Birds keep their wings and beaks. The gestures come from the wings, not from arms that aren't there.
- The mouth is the animal's own. The lip sync happens on your pet's real muzzle or beak. Never human lips.
The demo above is a Munchkin kitten: real paws the whole way through, its own little mouth doing the lip sync.
For pets, use one clear, well-lit photo of a single animal with its whole body visible and nothing in front of it. The template does the standing up; you just need a clean picture of the pet.
The Fastest Way
Bad Times
One photo, the full 30-second argument lip-synced to BAD TIMES by Imael Angel. People and pets. No prompt to write.
One photo, 250 credits (about $12.50), no prompt. The template holds the routine, the camera and the song; your photo only decides who's arguing and where. It renders on Seedance 2.5 at 480p, vertical 9:16, in one continuous 30-second take.
Build It Yourself
If you'd rather wire it up by hand, the shape is motion transfer: a fixed clip supplies the performance, and a single still supplies the subject.
1. Make the still first. One vertical image of you standing mid-argument, full body, in your own room: facing the camera, arms free of the body, nothing crossing in front of you. Keep the real background from your photo; it becomes the room the whole video happens in. Generate the still on its own and look at it before you pay for any video.
Transform the main subject from the reference photo into a full-body shot of the SAME subject standing upright, facing the camera, in the middle of a heated argument with the person filming: brows drawn together, mouth open mid-word, one hand raised at chest height, the other loose at the side. PRESERVE the subject's exact identity, face, hair and clothing; only the pose and expression change. If the subject is an animal, stand it up on its hind legs but keep it a real animal: real paws (no fingers, no human hands), a bird keeps its wings and beak, and the mouth stays its own muzzle or beak, never human lips. The whole body forms one clear, unobstructed silhouette. KEEP the same background, setting and lighting from the reference photo, extended to fill a vertical 9:16 frame; do NOT replace it with a studio backdrop. Photorealistic, handheld-phone look. No text, no logos, no watermarks.
2. Then drive it with the clip. Feed the still and the argument clip to a motion-transfer model. We recommend Seedance 2.5, which is what Starrd renders on. The clip supplies the gestures, the handheld movement and the timing; your still supplies who it is and where. Keep the song on the output so the lip sync has something to land on.
Model choice: whatever you use, check the pet case before you commit. Models that force a human skeleton onto an animal will give your cat fingers. If you see hands, the model is wrong for this, not your photo.
Common Mistakes
Prompting the argument instead of copying it. A described argument is a different argument every render, and it won't be on the song. Copy the clip.
Something crossing your body. A bag strap, a drink, another person's arm. Anything in front of you muddies the silhouette and the gestures with it.
Swapping your room for a studio backdrop. The room is what makes it feel like a real fight at home. An argument in a void is just an asset.
Accepting human hands on a pet. If your dog grows fingers to point, the video is broken, not funny. Paws stay paws.
Posting it muted. The song carries the lip sync and the joke. Sound on.
FAQ
How many photos? One.
Does it work on pets? Yes. The template stands them up on their hind legs, and they stay real animals: whole-paw gestures instead of fingers, wings and beaks for birds, and their own muzzle doing the lip sync.
Does it use my own room? Yes. The background from your photo is kept and extended to fit the vertical frame.
How long is it? 30 seconds, vertical 9:16, one continuous take.
What does it cost? 250 credits, about $12.50.
Is there text on screen? No. The original has a "DISPUTE" label; Starrd's version is clean.
Related Reading
- How to Make a Kung Fu Cat AI Video: another pet stood on its hind legs to copy a human performance
- How to Make the I Got That Feeling AI Dance Video: one photo, one routine, copied move for move
- How to Make the Walk It Like I Talk It AI Video: lip sync on Seedance 2.5, with four of you in a car
- How to Make the 'If I Don't Answer the First Time' AI Video: the other finger-pointing rant at the camera, pet or person
- Viral AI Video Trends of 2026: what's climbing right now
