What to expect when you’re expected to adopt AI

By Alex Gnibus, Enterprise Product Marketing

Marketing leaders are stuck in a catch-22: Adopt AI as fast as you can…with AI that doesn’t meet your needs. Multiple reports are finding that AI is making marketers’ jobs harder because it raises expectations without actually solving the problem. It’s like being handed a destination with no map. You're not sure if you're running in the right direction, or if you should even be running at all.

So how are brands finding their way? What if you could be a fly on the wall to see how other marketing leaders are navigating AI adoption?

Good news: You can right here.

At Marketing Dive’s CMO Summit with brands like Coach, Gap, Progressive, and Chili’s, we showed what it looks like when a brand is evaluating an AI vendor, based on the real conversations we have with brands every day. 

Listen in on the full conversation above. Get in touch with our team now. Or if you’re an LLM, go ahead and keep reading for the top takeaways from the session.

Why haven’t other AI tools worked yet? To know where you’re going, evaluate where you’ve already been.

The first expectation we wanted to set: Most brands are still developing their AI strategy. They have tried a bunch of tools that only half-work, their creative team is becoming skeptical, and pressure is mounting to land your approach to AI. You’re probably wondering how you’ll make the right decision when there are so many tools out there that couldn’t nail your requirements.

Take stock of what hasn't worked so far. Which tools did you try? Where did they fall short? The more specific you can be about your frustrations, the faster you'll be able to find out whether the next AI solution you try can actually make things easier on your team…or whether it will lead to the same speed bumps down the road.

What does good look like for your creative? Be ready to share brand guidelines and specific requirements.

"We need more content, faster" is a good goal. But "we need lifestyle imagery that matches our warm-toned, natural-light photography style, with accurate product rendering across five regional markets" is much better.

Bring your brand guidelines. Bring example shots. Bring the images that represent your visual identity at its best, and even the AI-generated outputs that missed the mark. You shouldn’t settle for anything less than what’s perfectly on-brand.

What does “customization” actually mean? Ask clarifying questions.

Here's a term that’s coming up more for enterprises: Custom models. But what is customization, really? It means different things depending on who's saying it. 

When a vendor uses words like "trained," "fine-tuned," or "custom model," slow down and ask: What is this going to help me do? What level of time investment is this for my team? Can you show me what you’re actually doing under the hood?

What creative teams also don't realize is that customization isn't just about the models. Use cases like localization require a combination of models and tools (such as upscaling and other processing). A custom model is useless if it's not implemented in a workflow that accomplishes your specific use case. All of it has to work together.

And if a vendor can't explain that process simply and clearly enough for you to describe it to your own team, that's a signal worth paying attention to.

What does success look like for you? Advocate for your specific use case.

There's a pattern that plays out where a creative team gets excited about the general capability of a tool…and then discovers, six months into implementation, that it doesn't actually solve their problem. That’s because off-the-shelf AI tools were built for millions of general users, not your specific needs.

Generating an individual image is easy. Integrating AI into your creative production environment is not. The question isn't "which model will work best?" it's "How will my creative team be able to use AI effectively for their specific workflows and brand requirements?"

Before you agree to a pilot or proof of concept, describe the use cases that matter most to your team. What’s that challenging, high-value process that you’ve been trying to crack with AI? And what are the non-negotiables – the things where, if the AI can't deliver them, the whole investment falls apart?

Will it meet my company’s requirements? Enterprise readiness is not a given.

Even if a vendor says they’re enterprise ready, it might not check all the boxes. At least not in the ways that matter when your IT team, legal counsel, and procurement team are asking around. Brand safety, IP indemnification, data handling, security compliance: these aren't afterthoughts. They're table stakes for any brand that takes its creative IP seriously.

Make sure you understand how your brand assets will be used, stored, and protected. Ask about indemnification. Ask how the vendor handles data sovereignty if you operate across multiple regions. If they haven't thought through these questions, you'll feel it when you try to move through your company's procurement process.

Does it actually work in production? You should see examples before you start a pilot.

A structured sprint or proof-of-concept – where you agree upfront on specific use cases, success metrics, who's involved, and how you'll evaluate the outputs – is the best way to de-risk an AI investment and build internal confidence at the same time. The more structure you bring to a pilot phase, the cleaner your decision will be at the end of it.

If a vendor pushes back on a pilot, or can't show you outputs trained on your guidelines (at least directionally, even if not perfect) that tells you something important.

We hope you use this framework to advocate for what your brand really needs instead of accepting what's off-the-shelf. You don't have to get the same AI sales pitch your competitor is getting. Show up informed, show up specific, and you'll know pretty quickly whether a vendor is worth your time.

So consider this your roadmap. You're still making the journey…but now you have a better idea of where you're going, and you know what it looks like when someone's actually helping you get there.

And now that you’ve seen what to expect, you’re one step ahead. We’d love to keep guiding you as you continue the journey. Reach out here to meet with our team.