Pollo AI turns creative ideas into campaigns with OpenAI

Pollo AI⁠(opens in a new window) was founded on the premise that people should be able to turn their creative ideas into finished images and videos without needing production expertise. Even with generative AI, creators and marketers can struggle to navigate different tools and choose between models to get the result they want. Some may even abandon their projects in frustration.

“We want to do for video what Canva did for design. We want to make it easy enough that more people can actually use it, not just the ones who already know how,” explains Bill Zhu, the founder and CEO of Pollo AI. Today, Pollo AI serves more than 26 million users who use the platform to create videos, ads, and visual content.

“As OpenAI’s models have improved, more of the workflows Pollo wanted to support have become practical inside the product. That means users can start from simpler inputs, spend less time figuring out the process, and get to stronger outputs more reliably.”

—Bill Zhu, Founder and CEO, Pollo AI

The team’s first breakthrough was video templates. Now, 30% of creator sessions start from these templates or trending-format workflows. Building on this success, the company’s latest offering is Pollo Agent. It uses OpenAI models, including GPT‑6 Astra and GPT‑Image‑2.5, to provide a fully guided workflow for video creation, including this post-apocalyptic adventure from a Pollo AI user.

Turning the user’s creative intent into finished content

Pollo Agent interprets the user’s rough ideas and visual references, and then develops storylines, scene flows, and scripts. CTO Peter Zhou says OpenAI models stood out for their ability to understand ambiguous creative inputs and their multimodal capability, sharing: “GPT‑5.6 performed especially well there.” The team also said they value the OpenAI API’s stability and release cadence.

—Peter Zhou, CTO, Pollo AI