Over the past year or so, I’ve noticed something funny happening in organizations. People who never would have opened Photoshop, written a campaign brief or drafted a customer-facing sales piece are now creating “marketing materials” in AI tools and sending them to the marketing team to be cleaned up, branded or finalized.

In some ways, this is a good thing. People are thinking about communication. They are trying to solve problems. They are taking initiative. And to be fair, some of what they produce is not bad. AI can help someone get their thoughts out, organize ideas and create a starting point much faster than staring at a blank page.

But a starting point is not the same as finished marketing, and at Radiant, this is where we’re starting to see many organizations run into trouble. Someone creates a draft in five minutes and assumes the remaining work should also take five minutes. After all, if AI was able to produce the piece that quickly, why can’t the marketing team simply “make it look nice” and send it out?

The problem is that marketing has never been just about making things look nice. Good marketing involves judgment. It considers the audience, the offer, the timing, the message, the tone, the ministry goal and the credibility of what is being said. It asks whether the piece is true, useful, clear, on-brand and appropriate for the situation. It pays attention to what is emphasized, what is left out and what the audience may misunderstand.

AI can produce content quickly, but it rarely knows whether that content is wise. That responsibility still belongs to people.

For ministries that care about their brand, this is an important moment. The question is not whether your team should be allowed to use AI. That horse has left the barn. People are already using it, and in many cases, they should be. The better question is how AI should fit into the marketing process without creating confusion, extra work or a lower standard.

I think the first step is to be honest about what AI is good at. AI can be very helpful for brainstorming, outlining, summarizing, rewriting rough thoughts and helping people get unstuck. It can give a fundraising person a better starting point. It can help a subject matter expert turn scattered notes into something more coherent. It can help internal teams think through ideas before they bring them to marketing. Those are useful gains.

But AI should not be treated as the final authority on anything that represents the ministry publicly. It should not decide the message. It should not invent claims. It should not bypass approval. It should not create a dozen different voices for the same brand. And it should not turn the marketing department into a cleanup crew for half-formed ideas.

That last point is worth considering carefully. When someone sends over an AI-generated piece and says, “Can you brand this?” they may think they are being efficient. But if the piece doesn’t include the audience, the objective, the source facts, the approval path or the intended use, marketing now has to work backwards. What was this meant to do? Who is it for? Is it accurate? Does it support the fundraising process? Has anyone with authority approved the claims? Is this even the right format? At that point, marketing is not polishing a finished piece. Marketing is trying to recover the missing strategy.

A better approach is to set some simple expectations. If someone uses AI to create a draft, that draft should come with a brief. It does not need to be complicated, but it should answer the basics: Who is this for? What are we trying to accomplish? What is the main message? What action do we want the audience to take? What source material was used? Who has reviewed the facts? When is it needed?

Those questions may feel obvious, but they prevent a lot of waste. They also remind the organization that speed at the beginning of the process does not remove the need for care at the end. It may also help to create a few clear categories for AI use.

For early thinking, AI can be used fairly freely. Brainstorming, summaries, internal notes and rough drafts are all reasonable uses.

For internal communications, AI can help with a draft, but a person should still own the message and review it before it is shared.

For customer-facing work, marketing should be involved before the piece is too far along. Fundraising tools, brochures, web copy, ads, email campaigns and social content all carry the ministry’s name, so they need to go through the normal review process.

For anything technical, legal, safety-related or reputation-sensitive, AI output should be treated as unverified until the right person signs off.

None of this needs to become a heavy policy document that scares people away from using tools that are helpful. Most ministries do not need a 40-page AI manual for their marketing department. They need clear rules of engagement.

  • Use AI to help you think.
  • Use AI to help you draft.
  • Use AI to help you organize.
  • Do not use AI to skip judgment, review or accountability.

That is the line.

The ministries that handle this well will not be the ones that pretend AI is irrelevant. They also will not be the ones that let every AI-generated idea become a branded asset. They will be the ones that allow AI to improve the front end of the process while protecting the quality of what goes out the door.

Because in the end, the end-user does not know or care how the first draft was created. They experience the final piece as your ministry speaking to them. If it is unclear, careless, off-brand or wrong, that belongs to you.

AI can help your team move faster. Used properly, it can help good ideas surface sooner and make collaboration easier. But it still needs a process around it. Otherwise, it will not reduce the burden on marketing. It will add to it.

Here’s a simple framework that can make a significant difference:

  • First, treat anything created by AI as a draft.
  • Second, require AI-generated marketing requests to include a brief.
  • Third, make sure customer-facing work still goes through the normal marketing review process.
  • Fourth, have subject matter experts review anything technical, legal, safety-related or reputation-sensitive.
  • Fifth, make it clear who owns the final product before it is published.

I believe this kind of structure doesn’t work against AI; it makes it more useful. Again, the goal is not to slow everyone down. The goal is to help the organization benefit from AI without lowering the standard of the brand.

Need help?

Wondering how AI should fit into your marketing process?

The Radiant team can help you think through the right guardrails, workflows and automation opportunities for your organization so AI becomes a useful tool instead of a marketing and communications distraction. Reach out to us and let’s talk about how to make AI work for your team, your process and your brand.

Ray Majoran
Ray Majoran CEO

Ray is the CEO of Radiant, where he focuses on building culture, creativity, strategic partnerships, and innovative technology solutions.