Do museums need a technologist to use AI well?

No. The real question is where it helps. The hard part is knowing what you want it to do for people.

Should a museum hide that it used AI?

No. Say when AI helped, say who checked it, and make it easy for people to tell you when it’s wrong. The Cleveland Museum of Art labels every AI-assisted description and invites visitors to flag what’s off.

Which museums are using AI to connect with visitors?

Remuseum’s report, 5 Ways That AI Can Humanize Museums, highlights the Frye Art Museum, the Cleveland Museum of Art, and Crystal Bridges. None of those projects replaced people.

AI as a Tool for Connection

Key Takeaways

Treat AI as a tool in service of people, not only as a risk to manage.
You don’t need to be a technologist. Start with what you want the tool to do for people.
Meet audiences where they look for answers, including AI summaries, and lower the barriers to the content.
Say when AI helped, say who checked it, and make it easy for people to flag what’s wrong.
Start with one draining task or one place visitors get stuck, then share what you learn.

I’m still thinking about the keynote fireside chat at MAAM this year. Meme Omogbai and Monica Montgomery covered a lot of ground, but the part that stuck with me was how they talked about technology and AI—not as a threat to fear or a trend to chase, but as a tool in service of people.

A lot of the AI talk in the cultural sector starts and ends with risk. And while those risks are real, if that’s the only conversation we have, we miss the chance to shape how these tools actually show up in our work.

Don’t be intimidated—be intentional

AI can be a great tool, and you don’t need to be a technologist to use it well. The real question is where it helps: speeding up the work that eats into your time, stretching what a small team can take on, and deepening the connection between visitors and the stories we tell.

After years of building museum experiences, I’ve learned that the tech is rarely the hard part. The hard part is knowing what you want it to do for people. Start there, and AI becomes a lot less scary.

“The hard part is knowing what you want it to do for people.”

Reach audiences where they are

People don’t find information the way they used to: more of them get answers from AI summaries (AEO) before they ever click a museum website.

AEO

Answer engine optimization. More people now get answers from AI summaries before they ever click a museum website, so the information has to be findable there—not only on the site.

AI can help close that gap. It can turn scholarly text into plain language for many different readers to personalize a normally static experience. It can let people search a collection in their own words instead of the words of a curator with a doctorate and thirty years of experience. And it can meet visitors in the gallery, on their phones, or at home after the visit.

The goal isn’t more screens. It’s fewer barriers between people and the content.

Be open about how you use it

Trust is what museums run on. Visitors believe what we put on the wall, on the screen, on our websites. That’s exactly why AI use shouldn’t be hidden.

The Cleveland Museum of Art gets this right. Every AI-assisted description is labeled, and visitors are invited to flag what’s off. That turns the public into partners with a stake in keeping content accurate. It also keeps staff expertise in charge of the final word.

Say when AI helped. Say who checked it. Make it easy for people to tell you when it’s wrong.

What this looks like in practice

Remuseum’s new report, 5 Ways That AI Can Humanize Museums, confirms this. It shares real examples from museums using AI to bring people closer, not push them away. A few stood out:

The Frye Art Museum built a clear brand voice with staff and board input, then used it to guide its AI writing tools. The project cost about $7,500. Since then, repeat web traffic is up 20% and attendance is up 18%.
The Cleveland Museum of Art launched AI-assisted visual descriptions for more than 68,000 objects. Staff wrote about 1,500 by hand first to set the standard, and every AI description is labeled and open to public feedback.
Crystal Bridges built AskRosie, a voice-first art companion. They made it a web app, not a download, because typical museum app adoption sits at just 3–5%.

None of these replaced people. In each case, the museum started with its mission and asked how AI could serve it. The report is also honest that hard questions remain—around intellectual property, data, jobs, and the environment.

Where to start

You don’t need a big budget or new tools to make this happen. Pick one task that drains your team or one place visitors get stuck, and try a small test. Learn from it, share what you find.

What’s one place in your museum where AI could help you connect with people?