Fancy Cafe POV
What a chatbot is good at here
ChatGPT and similar assistants are great at explaining the trend, writing a caption or describing the scene in detail. Many can also make a still image of you in a concrete cafe. What you need for this trend is a 15-second clip with timing: someone on the table, someone walking in, the look down, the climb off. That's a job for a video model.
What a video model needs from you
General AI video tools turn a prompt (and sometimes a photo) into a clip. To get this trend you have to write every part: the room, the two small round steel tables, the low bench, who sits where, when the second person enters and how slowly the first one reacts. Small changes in the wording change the result, so expect a few tries.
The shortcut: a video maker built for this one trend
The fancy cafe video maker already has the scene, the timing and an original backing track set up, so you only add a photo and pick a role. It uses Seedance 2.0 Fast by ByteDance to make a 15-second, 720p vertical video in about 5 minutes. 1 video is $8.88, paid once.
If you still want to try a chatbot
Ask it for a detailed shot list of the scene and paste that into the video tool you use. Keep your photo front-facing and sharp, and ask for a vertical 9:16 clip around 15 seconds long.