Still, Code and Theory doesn’t attempt to replace the entire production ecosystem with Higgsfield.
“AI is still not at a stage where it’s perfectly indistinguishable from live action or real motion, both in physics, in human reaction and emotion,” Foster said. Legal risks, such as copyright infringement, also shape how agencies choose which AI tools to use: Platforms like Adobe Firefly, which are designed to be safer for commercial use, are often used to develop characters or brand assets early, while tools like Higgsfield are brought in later for creative control and developing campaigns faster.
A WPP executive who wasn’t authorized to speak to media told ADWEEK that Higgsfield appeals to teams without deep prompt‑engineering expertise. The platform’s stylistic presets reduce guesswork, allowing people to generate editorial‑style visuals—high‑contrast lighting, wide cinematic angles—without crafting complex prompts. In practice, the executive said, Higgsfield is often used internally or presented to clients as conceptual groundwork for larger campaigns rather than final output.
In internal testing, WPP often runs identical prompts across competing systems, finding that results vary based on how each model has been trained. The question becomes less about which model is seen as “the best” and more about which produces the right creative outcome with the least friction.
Zero production tax
For Mashrabov, that flexibility is the point. He isn’t interested in the AI slop currently clogging feeds—those repetitive, uncanny-valley ads that feel like cheap replicas of a mid-2000s infomercial. Instead, he’s betting that the next era of advertising will be won net‑new concepts that are only possible with AI, making it easier to localize and adapt content across demographics and languages.
“We decrease production tax to zero with AI,” Mashrabov said.
Looking ahead, Higgsfield has plans to expand into performance marketing, allowing performance data to feed back into the content generation process and guide which variations are worth producing.
In an increasingly crowded AI video market—one that has pulled in more than $500 million in new funding in 2025 alone—Mashrabov is betting that marketers don’t need another black box but a system that helps ideas gain mass fast enough to matter.
