AI Is Changing Marketing. The Interesting Part Is How Marketers Are Changing With It.
By James Arnold
Rooster Chief Digital Officer
Spend enough time talking about artificial intelligence and eventually the conversation gets pulled toward one of two extremes.
AI is going to change everything.
AI is overhyped.
After talking with a number of marketers about how they’re actually using AI, I’m increasingly convinced that neither is a particularly useful way to think about it. AI is changing marketing. In some areas, very quickly. But the more interesting story isn’t the technology. It’s about what marketers are learning to do with it.
That became one of the clearest themes in the conversations we had for the first episode of AMiDA 2026. We talked with people working across marketing, media, analytics, technology and agriculture. Their experiences with AI varied. So did the tools they use and the ways they use them. But there was a surprising amount of agreement about where all of this appears to be headed.
We’re moving beyond experimentation
Not very long ago, most conversations about generative AI started with ChatGPT and ended with something like “Have you played around with it yet?” That phase is ending.
The marketers we talked with aren't just experimenting anymore. They're beginning to incorporate AI into actual workflows. That matters.
Using AI to write a first draft is interesting. Using it to help analyze information, develop ideas, interrogate data, explore alternatives, accelerate research or work through a problem begins to change the way the work itself gets done. And that seems to be where some of the biggest opportunities are emerging.
The best applications aren't necessarily the flashiest ones. They're often the places where AI removes friction. A task that used to take hours might take 30 minutes. A marketer can explore five ideas instead of one. Someone who isn't a data scientist can ask better questions of a dataset. A team can get farther into an idea before bringing in additional resources.
None of those things sounds particularly revolutionary. But add them together, and marketing starts to operate differently.
The real skill may be knowing what to ask
One theme surfaced repeatedly in these conversations: AI tends to become more useful as the person using it becomes better at working with it. That sounds obvious, but it has some important implications.
We have spent a lot of time talking about “prompt engineering,” sometimes making it sound like there is a secret collection of magic words that unlocks AI. That’s really not the point. The bigger skill is knowing what you're trying to accomplish.
What information matters? What assumptions should be challenged? What context does the system need? What would a good answer look like? What should you ask next?
Those aren't AI skills as much as they are thinking skills. And that may be encouraging news for experienced marketers.
People who understand customers, markets, strategy, messaging and business problems already possess much of the knowledge needed to use these tools well. AI doesn't eliminate the value of that experience. In many cases, it amplifies it.
Give two people the same AI tool and you're unlikely to get the same result. The person who understands the problem better has an advantage.
Faster isn't the same thing as better
Of course, speed is one of AI's most obvious benefits. Things happen faster. Research happens faster. Drafts happen faster. Summaries happen faster. Ideas happen faster. That's valuable.
But there was also an important undercurrent in our conversations: efficiency can't be the only objective.
If AI simply helps us produce more mediocre content faster, I'm not sure we've accomplished much. The opportunity is to use the time we gain to improve the work. Explore another direction. Ask another question. Look at another source. Spend more time thinking about the audience. Push an idea further. That's a much more interesting proposition than simply finding ways to remove people from the process.
The organizations that benefit most from AI may not be the ones asking, “How many hours can we eliminate?” They may be the ones asking, “What can our people do now that they couldn't do before?”
Human judgment becomes more important, not less
There is an irony in all of this. The better AI gets at producing things, the more valuable judgment becomes.
AI can generate a lot of answers. Someone still has to decide whether they're any good or if they even answer the important questions. That requires context.
It requires knowing the customer and understanding the business. It requires recognizing when something technically correct doesn't feel right. It requires knowing when an idea is generic, when a claim needs verification, when a piece of creative misses the point or when the data doesn't tell the whole story.
Agricultural marketing makes this especially apparent. Agriculture is full of nuance. Producers aren't one homogeneous audience. A dairy in Wisconsin isn't a cow-calf operation in Kansas. A corn grower doesn't necessarily think about risk, technology or purchasing decisions the same way specialty crop producers do.
The language matters. The context matters. Credibility matters. AI can help us work with all of that information. But knowing what actually rings true still requires people who understand the market.
AI can make expertise more accessible
One of the possibilities I find most interesting is what happens when AI gives more people access to capabilities that previously required specialized expertise. That doesn't mean expertise goes away. Far from it! It means the starting line moves.
A strategist can do more initial data exploration. A writer can dig into research in ways that previously would have required hours of manual work. A small marketing team can explore creative directions that would once have required considerably more time and resources.
And the technology itself is beginning to show up in places beyond the chatbot, which is fantastic. It will increasingly become part of the experiences marketers create.
We're still early
Perhaps the easiest mistake to make right now is assuming we've already figured out what AI means for marketing. We haven't.
The tools are changing too quickly. Marketers are still experimenting. Organizations are still figuring out policies, workflows and expectations. And many of the applications that will eventually seem obvious probably haven't been invented yet.
That uncertainty can be uncomfortable. I think it is exciting.
There are moments in marketing when a technology arrives and everyone immediately starts looking for the new rulebook. Right now, there isn't one.
That gives marketers an unusual opportunity to help write it. We can experiment. We can figure out where AI genuinely improves the work and where it doesn't. We can determine which tasks should become automated and which ones deserve more human attention.
And we can learn what happens when talented people suddenly have access to capabilities that would have seemed improbable only a few years ago.
The marketers who stay curious will have an advantage
If there was one attitude connecting many of the people we interviewed, it was curiosity. They weren't waiting for someone to hand them the definitive AI strategy. They were trying things. Testing tools. Finding useful applications. Learning where the technology fails. Changing their workflows. And asking what else might be possible.
That seems like a pretty good model for the rest of us. You don't need to become an AI expert. You probably do need to become comfortable working with it. Because the important question is quickly becoming less about whether AI will have a role in marketing. It already does. The more useful question is what we're going to do with it. And based on the marketers we talked with, we're only beginning to find out.