Can ChatGPT or Gemini make the Bulin 47 video?

ChatGPT, Gemini and the real clip side by side

ChatGPT and Gemini facts as of 15 October 2026; both change often, so check the app before you plan around them.

What ChatGPT can do for the trend

Ask ChatGPT for "the Bulin 47 video" and it may offer to make it, but what comes back is a picture. That picture still has uses: a cover image for your post, or a full-body still of you to feed a video app that needs one.

Two tips. Leave real names out of the request; image tools often refuse or blur a named person. And describe the scene instead: night, a red-lit doorway, a portable speaker, a phone held at chest height. Our Bulin 47 AI prompt has a ready one.

What Gemini can do

Google's video page for Gemini says Gemini Omni has replaced Veo in the Gemini app, makes 10-second videos with sound, can start from up to five photos, needs a Google AI subscription, and that features vary by plan and country (18+). Every video carries Google's SynthID watermark.

• You get a new street that looks like the clip, not Kevin Wai's recording.
• The sound is the model's own. Add the freestyle sound when you post.
• Ten seconds is shorter than the moment most people copy.

Why a chatbot cannot copy the real clip

A text-to-video model draws from what you describe; it has no copy of Kevin Wai's footage to work from. The versions in your feed that follow Bulin 47's every gesture come from tools that take the clip itself as a reference and redraw the performer. The methods are compared in how the Bulin 47 AI trend works.

YouTube

From one photo to the real freestyle

On the Bulin 47 video maker there is nothing to prompt: add one photo, and the performer in the real clip is redrawn as you, with the original sound laid back on. You get a file with sound and a silent copy. Each Google account can start with a free 2-second preview.