OpenAI announced the tool as a way to create detailed images and cinematic video advertising with three named models.
OpenAI says Pollo AI helps creators turn bold ideas into detailed images and cinematic video ads. It associates the tool with GPT-5.6, GPT-6 Astra, and GPT-Image-2.5.
That points to a creative-production tool: users bring an advertising concept, and Pollo AI is presented as helping turn it into visual material. OpenAI’s description does not explain how the three models work together, what each model does, or how much control users have over the output.
The announcement establishes a claimed capability, not campaign results. It does not show whether the generated ads improve sales, reduce production costs, or outperform other tools. Those questions would require measured results from real campaigns or independent testing.
Access is also important for teams considering the product. The supplied announcement does not establish whether Pollo AI is generally available, limited to selected users, or offered to paying customers. It also does not establish whether every user can access all three named models.
For now, potential users should check the product’s access terms and try a real campaign brief if access is available. The practical test is whether the tool can produce usable images and video while following a brand’s visual rules and creative direction.