Using AI in Brand and Design Work: 10 Mistakes to Avoid

Half-human, half-robot face with exposed circuitry

AI is powerful. It is not a strategy.

The question for most organizations is no longer whether to use AI in communications and design. Your team is probably already using it. The real question is how to use it without losing your voice, your audience's trust, or your judgment.

I use AI in my own work every week. It helps me explore variations, test palettes, and mock up ideas faster. It does not replace intent, or the ability to listen, synthesize, and decide. This guide covers the ten most common mistakes I see organizations make with AI, and practical guardrails for getting it right.

Assistant, not artist

The most useful mindset shift is simple. Treat AI as an assistant, not the artist.

An assistant can sketch twenty quick versions of an idea in minutes. That's genuinely valuable. It frees time to refine the concept, focus on the story, and make better decisions. Generation without judgment, though, is not creativity.

The danger comes when AI becomes the artist. That's when we get soulless logos, templated identities, and visuals that don't fit the audience. The story and the strategy still have to come from people who understand the mission and the audience.

Clients sometimes worry that AI will push designers aside. In my experience, the opposite is true when it's used well. I'm not handing off the emotional core of a project. I'm using AI to get there faster, with more room for iteration and empathy. The human connection still shapes the final product.

Ten mistakes organizations make with AI

1. Automating without purpose

Using AI because it's available, not because it solves a problem. Start with the bottleneck. Is it first drafts? Resizing assets? Summarizing research? Apply AI where it removes real friction.

2. Ignoring ethics

AI image and text tools raise real questions about consent, ownership, and representation. For nonprofits and public agencies, using AI-generated images of "community members" who don't exist can undermine the trust your work depends on.

3. Replacing judgment

AI can generate options. It can't tell you which one is right for a grieving family, a skeptical community, or a nervous first-time donor. Decisions that require understanding people stay with people.

4. Moving faster than trust

Your audience's comfort with AI may be lower than your team's. Rolling out AI chatbots, synthetic spokespeople, or generated imagery without considering how people will react can damage relationships you spent years building.

5. Skipping governance

Without clear guidelines, every staff member makes their own rules. That leads to inconsistent quality, accidental disclosure of sensitive information, and content no one reviewed. Many government agencies now have formal AI policies. Nonprofits need them too.

6. Overpromising outcomes

AI makes production faster. It doesn't make strategy better, and it doesn't fix a weak message. Expect efficiency gains, not miracles.

7. Underestimating risk

Generated content can include factual errors, biased assumptions, or material that resembles someone else's copyrighted work. In public health and government communications, an error isn't just embarrassing. It can mislead people about their health or their benefits.

8. Forgetting accessibility

AI-generated alt text is often vague or wrong. AI-generated designs often fail contrast standards. Automated captions miss words and names. Every AI output still needs an accessibility check. My Section 508 and color contrast guide covers what to look for.

9. Losing brand voice

AI writing tools default to a generic, slightly enthusiastic tone. Left unchecked, your newsletter, website, and social posts start sounding like everyone else's. Your voice is one of your most distinctive assets. Protect it.

10. Treating AI as magic

AI is a tool with real strengths and real limits. Teams that understand both use it well. Teams that treat it as magic are surprised when it fails.

Where AI helps in design and communications

Used well, AI is a strong assistant for:

  • Exploration. Generating rough variations, layouts, and directions to react to

  • First drafts. Starting points for copy that a writer then shapes

  • Production tasks. Resizing assets for different platforms, background removal, simple retouching

  • Research synthesis. Summarizing survey responses or interview notes for a human to review

  • Internal communication. Turning dense information into clearer visuals and summaries

That last one is growing fast. At SXSW London, a recurring theme was that visual tools like Canva and Flourish are becoming essential for internal communications, not just marketing. Younger audiences expect clear visuals that make ideas easier to grasp, not text-heavy decks.

Where humans must decide

  • Brand strategy and positioning

  • Final creative direction

  • Anything representing real people or communities

  • Messages about health, safety, legal rights, or benefits

  • Final review of every piece before it's published

Guardrails for using AI well

A simple AI policy for a communications team can fit on one page. It should cover:

  1. Approved tools. Which tools staff can use, and for what.

  2. What never goes in. Personal information about clients, donors, or patients. Confidential documents. Anything you wouldn't post publicly.

  3. Human review. Every AI-assisted piece is reviewed by a person before publishing.

  4. Image rules. When AI imagery is acceptable, and when only real photography will do.

  5. Disclosure. When and how you tell your audience AI was involved.

  6. Voice check. AI drafts are edited against your voice guidelines.

  7. Accessibility check. Alt text, captions, and contrast reviewed by a person.

  8. Ownership. Who maintains the policy and updates it as tools change.

The goal isn't to slow your team down. It's to let them use AI confidently, knowing where the lines are.

How to start: a 30-day pilot

If your team is using AI informally, a short pilot turns scattered experiments into something you can evaluate.

Week 1: Pick one workflow. Choose a single, repeatable task with a clear bottleneck, such as resizing campaign graphics for every platform or producing first drafts of event announcements.

Week 2: Set the guardrails. Write the one-page policy above, even in rough form. Decide who reviews output and what "good enough" means.

Week 3: Run it. Use AI for that workflow only. Track the time it saves and every correction a person had to make.

Week 4: Review honestly. Did quality hold? Did it sound like you? Did it save meaningful time after review? Keep what worked, adjust the policy, and pick the next workflow.

This keeps AI tied to real problems, gives leadership evidence instead of hype, and builds the habits that protect your brand as use grows.

Frequently asked questions

Should nonprofits use AI-generated images?

Be cautious. For illustrations and abstract graphics, AI can be useful. For images that appear to show your real community, real photography with consent builds far more trust.

Can AI design our logo?

AI can generate logo-like images, but a logo is the result of strategy: who you are, who you serve, and how you want to be seen. AI can help explore directions. It can't make those decisions.

How do we keep our brand voice when using AI writing tools?

Write clear voice guidelines, include examples, and have a person edit every draft against them. Treat AI output as a rough first draft, never a final.

A thoughtful approach to AI and design

Design has always been about communication between people. Tools change. The goal doesn't. If your organization wants to use AI without losing its voice, I help leadership teams set guardrails that keep empathy and brand intent in charge. Learn about my executive creative leadership work, or let's connect.

Related guides: What a fractional creative director does ยท Branding principles that last

David Ostroff

We are a full-service design agency that provides dynamic solutions for financial, government, non-profit, commercial and arts organizations.

https://www.davidostroff.com
Previous
Previous

Why Strategy Comes Before Design, and What a Strategy Phase Looks Like

Next
Next

What a Fractional Creative Director Does, and When to Hire One