From Image Tool to AI Director

Diving deeper into

Coco Mao, CEO of OpenArt, on building the TikTok for AI video

Interview
We don't want users to diligently write prompts for each frame and then convert them into videos.
Analyzed 5 sources

OpenArt is trying to move the product from an AI tool into an AI director. The key idea is that the real user job is not making single images, it is turning a rough story into a finished piece of content. That means automating script writing, storyboarding, character consistency, clip generation, audio, and assembly, so creators and SMBs can get to a publishable video without learning a complex frame by frame workflow.

  • The current workflow is still stitched together by hand. Creators often start with a script, generate key images, turn those into clips, then combine everything with sound and edits. OpenArt is building around that bottleneck, especially the hard problem of keeping the same character and style coherent across scenes.
  • This is also how OpenArt separates itself from model labs like Runway and Sora. Those products give more manual control for power users and filmmakers. OpenArt is aiming at the much bigger group that wants a push button result, similar to how Canva hid Photoshop complexity behind templates and simple controls.
  • The strategy matters because image generation alone became crowded fast. OpenArt grew from about $1M ARR in 2023 to $10.5M in 2024 and $70M in 2025, while shifting from standalone image creation toward video storytelling. The revenue acceleration suggests workflow packaging is becoming more valuable than the underlying model itself.

From here, AI video will keep splitting in two. One lane will serve experts who want shot level control. The other will serve creators, marketers, and small teams who want to type an idea and receive a finished asset. OpenArt is betting the larger opportunity sits in owning that second lane, then expanding from tool into distribution and eventually an AI native content platform.