KNEECAP (2024), directed by Rich Peppiatt

So far, we have seen AI become an indispensable tool for advertising agencies in multiple ways. As a treatment maker, the way it comes across directly my way is through animatics, the short video agencies create to give their clients and production companies a better idea of what the film looks like and its flow. Again and again, I have seen animatics made by AI, when they used to be made by animators and illustrators, or by editors editing together a range of clips from existing commercials to come up with a base for a brand new commercial.

As these tools become more accessible, a new question is surfacing. What happens when the pitch looks so complete that the client begins to ask whether the film even needs to be made?

This is the illusion of finish. And it is becoming a central concern in pitch work.

When a polished sequence appears midway through a deck, it can distort the client’s perception. It may trigger the impression that the idea is already resolved. The treatment stops feeling like a proposal and begins to resemble a finished product. This shift has implications that go far beyond aesthetics. It changes how the work is valued.

Treatments do not simply communicate ideas. They shape how those ideas are received — and, perhaps more importantly, invite the client into the process. A strong treatment makes room for the client to contribute, to shape the proposal in ways they believe will serve their audience. And anyone who has worked in advertising knows that clients often want some degree of control, if not complete authorship. The treatment is their first entry point into that process.

Seeing something that looks “more or less finished” — especially when it has been generated by AI — can carry unintended consequences.

First, it may come across as passive work. A deck that feels like it was assembled by prompting a pre-existing script can suggest that the director is simply echoing the agency’s vision, rather than advancing it. It risks making the agency wonder why they should hire someone at all, when the work could have been executed internally. In that sense, the director’s opportunity to bring original thinking to the table is lost.

And that is precisely where the value of the director lies. Now more than ever, it is time to bring forward the distinctive qualities that only a director can offer — a specific way of seeing, a visual vocabulary, a lived point of view. These are things AI can support, but not originate. They emerge from experience, memory, intuition, and taste — from the director’s accumulation of reference and meaning over time.

The second implication touches on something we have written about before. Often, the scripts sent over by agencies are not fully formed. They arrive as creative outlines, asking to be shaped. This puts pressure on the director and their team to elevate the idea, to give it rhythm, depth, atmosphere. And when that’s the task at hand, leaning too heavily on generative tools to do the imagining can result in something that feels less crafted, not more.

AI cannot fill gaps in creative thinking. It can only follow the shape that is already there.

This is not a critique of the tools themselves. It is a call to think more carefully about how they are used and what they suggest. An AI generated image may be striking. It may even help a concept work.

But if it implies that the idea is already fully formed, it risks undermining the sense of authorship — and authorship is what differentiates one treatment from another. It is the reason one director is chosen over someone else.

I know many would suggest that, at its core, a treatment is just a combination of references pulled from other sources, and there is absolutely a point in that. The difference here is that we know what the source is, and we actively avoid replicating it completely. These references are meant to serve as visual anchors, helping the client understand the tone or idea being proposed. They are not meant to become the base from which the director copies. The goal is to translate a vision, not to reproduce someone else’s.

That brings us to the question of intentionality. Not every high resolution image improves a treatment. Not every sequence adds clarity. Sometimes a single, imperfect still from an old film can communicate tone or movement more precisely than an entire AI rendered composite. The question is not about polish. It is about control, about knowing what to show, when to hold back, and how to keep the audience engaged in the idea rather than distracted by the output.

At Ghost, we find ourselves hoping that the saturation of perfect, frictionless imagery might open space for a new visual trend: one that embraces the unexpected, the rough edged, the atmospheric. One that moves away from golden hour sunsets and default beauty, and toward something more idiosyncratic and emotionally real.

As treatments evolve alongside new technologies, the challenge is not to keep up with every new tool for its own sake. It is to use them with clarity and purpose, in ways that preserve the space between concept and delivery. That space is where curiosity grows.

And that conversation is what makes the production worth pursuing in the first place.

What Next?

If this resonates with you, we’ll be sharing more deep dives into the craft of treatment writing and design. Let us know if there’s a topic you’d like us to explore next.

🔗 Check our work at http://www.treatmentsbyghost.com
🔗 Job inquiries info@treatmentsbyghost.com
🔗 Follow @ghost_treatments for more insights