Making realistic AI video still takes more than better prompts
A video team tried for two days to recreate a highly realistic AI video and could not match it. The target video showed small natural actions before the person spoke, including hair adjustment, quick side glances, micro-expressions, a hurried walk, smooth turns between clips, and believable people in the .
The only clear signs of AI were small mistakes, such as a tattoo appearing and disappearing. Flow was tested, Claude Code was used to write prompts, and Gemini was given the example video to suggest prompts.
The character was also set up inside Flow with those details added to the character instructions. The used image-to-video and connected clips by using the last frame of one clip as the seed for the next, but the hard part remained the small movements inside each clip.
Key points
- The hardest part was reproducing small natural actions inside each clip.
- Flow, Claude Code, and Gemini were all tried without the result.
- The used image-to-video plus last-frame linking between clips.
- Small s, such as a tattoo changing, still exposed the video as AI-made.
- Character instructions did not fully control micro-expressions and realism.