Quick answer
Starrd's I Be Thinking Bout My Bills template recreates the viral 'I be thinking bout my bills' restaurant-counter meme from one to three photos. Photo 1 is the one talking, who laughs into the camera about thinking about their bills, then about somebody else's, and leans over the counter to explain. Photo 2 is the one in the middle, who keeps asking "You talking about me?". Photo 3 is the one at the end, who watches the whole time with a phone in hand and never says a word, until the punchline lands on them: "I'm talking about him." Any empty seat is filled with an AI-generated stranger. It is one continuous handheld take that only pans, with every line lip-synced to the original audio, which is uncensored and contains explicit language. People and pets, 18 seconds, widescreen 16:9, Seedance 2.5, 150 credits (about $7.50).
One to three photos in, and out comes you and your friends at a restaurant counter recreating the viral "I be thinking bout my bills" meme: one of you laughs into the camera about bills, the one in the middle keeps asking "You talking about me?", and the punchline lands on the one quietly listening at the end. One take, every line lip-synced, people or pets. Here it is:
I Be Thinking Bout My Bills
One to three photos. One of you laughs about bills, one keeps asking 'you talking about me?', and the joke lands on the quiet one at the end. 18 seconds, widescreen.
What You Get
- One take, no cuts. A low, handheld vlog shot from the near end of the counter that only pans: right and closer at about 3 seconds, left onto the other two at about 14.5.
- Three seats. Photo 1 talks, photo 2 sits in the middle, photo 3 sits at the end. Fewer photos? The empty seats get AI-generated strangers.
- Real lip sync. Every line is mouthed to the original audio, and the one at the end never says a word.
- The performances. The grin straight into the lens and the hand up beside the face, the middle seat's offended hand on the chest, and the deadpan stare from the end.
- People, pets or both, in any seat.
- 18 seconds, widescreen 16:9.
What the Meme Actually Is
The clip was posted by @kobilol0 on November 1, 2025, with the on-screen caption "You talking bout me" 😭, and has about 1.01 million plays, 88,000 likes and 41,000 shares (as of October 16, 2026). Three people sit in a row at a restaurant counter at night, filmed handheld from the near end of the counter.
The one nearest the camera is laughing about money: he doesn't know how much the other guy has, but "I be thinking about my bills", and then he's thinking about the other guy's bills too. The one in the middle assumes it's about him and asks, twice. It isn't. The whole time, the one at the end has been sitting there with his phone, listening, straight-faced.
Why it suits AI: you need three people, a counter and timing that lands to the half-second. One to three photos fill the seats, and the timing comes with it.
The Beats, Line by Line
| Time | Who | What happens |
|---|---|---|
| 0:00 | The one talking | Laughing at the near end of the counter: "I ain't gonna lie…" |
| 3.5s | The one talking, grinning into the lens | "…I be thinking about my bills…" and then about the other guy's bills |
| 6.4s | The one talking | A hand up beside the face, still laughing |
| 7.5s | The one talking, turned side-on to the middle seat | Leans over the counter with open hands: "I said this before, bro. It's real…" |
| 10.45s | The one in the middle | "You talking about me?" |
| 10.85s | The one talking | "Nah, him." Then wonders out loud how that guy makes money |
| 12.8s | The one talking | Ducks behind a hand, cracking up |
| 13.55s | The one in the middle | Again: "You talking about me?" |
| 14.4s | The camera | The one talking leans out of frame right, and the camera pans left onto the other two |
| 15.0s | The one talking, off-screen | An explicit comeback, while the one in the middle, offended, puts a hand flat on the chest |
| 16.1s | The one talking, just an arm | Reaches back in, pointing past the middle seat at the end: "I'm talking about him." |
| 16.9s | The one in the middle | Cracks up. The one at the end, phone in hand, still hasn't said a word |
Who sits where: photo 1 is always the one talking, nearest the camera on the right. Photo 2 sits in the middle, photo 3 at the end. Put the friend who laughs at their own jokes first.
Pets Welcome
- Their own mouth. A pet talks with its own mouth, on the same timing as the original lines.
- Their own face, fur and real paws. No human hands on animals.
- Any seat, any mix. People and pets sit side by side. If every photo you upload is an animal, the strangers in the empty seats are animals too. Demo renders include a golden retriever, a ginger tabby and a raccoon.
Your Look, Their Counter
- Your own face, hair and outfit. Nobody gets the original's cap, chain or watch.
- Their counter. Three seats in a row along a long white counter, a light-wood pegboard wall with a red-and-white striped flag artwork on the left, and a dark back of the room with warm pin-spot lights, at night.
- Or your place. Name somewhere else in the Director's Notes, like a sushi bar or a diner, and the scene moves there, still with three seats along a counter, bar or table.
This one has explicit language. The audio is the original sound, uncensored, including the n-word and the f-word, and none of it is bleeped. Keep that in mind before you post or share it.
Upload the speaker first: photo 1 always does the talking and is on screen for most of the take. The one at the end barely moves, so any photo works there. Short a friend? Leave a seat empty and Starrd casts a stranger in it.
The Fastest Way
I Be Thinking Bout My Bills
One to three photos, one 18-second take at the counter with the original audio, every line lip-synced. People and pets.
One to three photos, 150 credits (about $7.50), no prompt. It renders on Seedance 2.5 at 480p, widescreen 16:9.
Build It Yourself
1. Make a three-seat storyboard sheet. Generate the counter from where the camera sits, low at counter height at the near end, with all three people in their seats: photo 1 nearest the camera on the right, photo 2 in the middle, photo 3 at the far end with a phone in hand. Add a panel for the speaker grinning into the lens and one for the other two after the pan left, so the model has seen every framing the take passes through.
2. Write it as one take with a beat list. A model left to its own devices loves to cut to close-ups. Say it's one continuous handheld shot that only pans, and give every beat a time.
3. Time the mouths. Write exactly when each person's mouth moves, so the middle seat stays still while the speaker talks, and the end seat never moves at all.
One continuous handheld vlog take, never a cut, low at counter height from the near end of a long white counter at night. Three people sit in a row: the person from reference image 1 nearest the camera on the right, the person from reference image 2 in the middle, the person from reference image 3 at the far end, holding a phone. 0 to 10.5s: person 1 laughs and talks; at about 3s the camera pans right and closer; from 3.5s to 6.3s person 1 grins straight into the lens; from 6.4s to 7.5s a hand goes up beside the face; then person 1 turns side-on and leans over the counter toward person 2, talking with open hands. 10.45 to 10.85s: only person 2's mouth moves, asking a short question. 10.85 to 12.8s: person 1 answers. 12.8 to 14.4s: person 1 ducks behind a hand, laughing; from 13.55s to 14.3s person 2 asks the same question again. 14.4s: person 1 leans out of frame right and the camera pans left onto persons 2 and 3. 15.0 to 16.1s: person 1 talks off-screen; person 2, offended, presses a hand flat to the chest. 16.1s: person 1's arm reaches back into frame, pointing past person 2 at person 3. 16.9s: person 2 cracks up. Person 3 watches person 1 the whole take, straight-faced, mouth closed, and never reacts.
4. Render with the audio. Feed the sheet, the photos and the original audio to a model that takes reference images and audio (we recommend Seedance 2.0 or 2.5). Lay the original audio back on at the end so no line gets dropped.
Common Mistakes
Cutting between people. The joke works because the camera has to swing around to find the one at the end. Close-ups give it away early.
No speaking windows. Without them, whoever is on screen mouths along to everyone else's lines.
A reactive end seat. The one at the end has to stay deadpan the entire time, even while being pointed at. Say so, or they laugh along.
Describing the original people. Describe everyone only from their own photo, and name the cap, chain and watch as things nobody wears.
Pet rules on the wrong subject. Write the own-mouth, real-paws rules only for the seats that are animals, or the people get paws.
FAQ
How many photos? One to three. Photo 1 always talks; any empty seat gets an AI-generated stranger.
People or pets? Both, in any seat, mixed casts included.
Is the language explicit? Yes. The original audio is uncensored.
How long is it? 18 seconds, widescreen 16:9.
What does it cost? 150 credits, about $7.50.
Related Reading
- How to Make the 'Like 700 Mile Per Hour' AI Interview Meme: another viral clip, every line lip-synced
- How to Make the 'Yo Dookie, Pick Up The Phone' Wazzup AI Video: another meme for you and your friends
- How to Make the 'Tiptoeing in My Jordans' AI Video: another viral clip, people or pets
- Viral AI Video Trends 2026: every trend we've turned into a template this year
