From Generation to Storytelling Workflows

Diving deeper into

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

Interview
the market has been saturated by tools with increasingly similar capabilities
Analyzed 5 sources

Saturation means raw generation is no longer where durable value sits, because most image tools can now make broadly similar outputs and the real product battle has moved up the stack into workflow, audience, and packaged use case. OpenArt grew quickly in image generation, but the company now frames its wedge around turning a rough idea into a finished story, with script help, storyboard frames, character consistency, image to video, and audio assembled into one simpler flow for creators and SMBs.

  • OpenArt describes its own earlier image product as close in core capability to Midjourney and Ideogram, which is why it is shifting from a general AI art tool into visual storytelling. The product aim is less make an image, more make a complete social video or short narrative from a single prompt.
  • This split is showing up across the market. Runway is going deeper for filmmakers and VFX teams with shot level control and its own models, while OpenArt and similar product companies are bundling open models into easier creator workflows. The difference is who does the work, the user or the software.
  • As features converge, winners increasingly look like aggregators or workflow products, not standalone model wrappers. Higgsfield’s jump to an estimated $230M ARR by January 2026 came from packaging multiple models behind presets and ad creation workflows, not from winning on a single underlying model alone.

Going forward, the category should keep compressing at the model layer and expanding at the workflow layer. The strongest products will be the ones that own repeatable creative jobs, social ads, story videos, product clips, and character based content, while swapping in better models underneath as they appear.