3 Things Cultural Institutions Should Consider Before Using AI

Welcome in.

This is my first blog post, so it feels only right to start with a subject that’s equal parts thrilling and thorny: what happens when cultural institutions — our museums, libraries, galleries, and archives — start inviting AI into the fold?

We’re in a moment where artificial intelligence is being sold as both oracle and oracle-maker. The promise? Faster processes, deeper insights, broader reach. But for institutions that hold stories, memory, and meaning — especially those charged with representing the fullness of humanity — the decision to use AI isn’t just about innovation. It’s about intention.

Before you jump headfirst into the algorithmic deep end, here are three things I believe cultural orgs should sit with, question deeply, and get aligned on.

1. Start With the “Why”: What are you trying to fix — or reimagine?

AI is seductive. It dazzles. But just because something is possible doesn’t mean it’s purposeful.

Too often, cultural orgs are tempted to adopt AI because it’s trendy, not because it serves a deeper goal. But real transformation starts with clarity.

• Ask yourself: What challenge are you facing that AI could actually solve? Maybe you’ve got a mountain of uncatalogued material. Maybe visitors keep asking the same questions, and you’re wondering if a chatbot could handle the FAQs while your team digs into the nuanced stuff. Maybe you’re dreaming up ways to make your collections more accessible or more emotionally resonant.
• Anchor it in your mission: How does AI align with your values? If your institution is rooted in inclusion, equity, or decolonial practice — how does this tech support (rather than undermine) that?
• Start with a small, intentional experiment: Don’t aim for a system overhaul on day one. Try a pilot — a thoughtful test case that’s measurable, low-risk, and capable of teaching you something meaningful.

Prompt to ponder: What would it look like if AI helped your team breathe easier, your collections speak louder, or your visitors feel more seen?

2. Interrogate the Ethics: Who’s reflected — and who’s erased?

Let’s not sugar-coat it: AI inherits the biases of the world it’s trained on. And if we’re not careful, it’ll replicate those blind spots in ways that feel slick, objective, and “neutral” — when it’s anything but.

• Biased in, biased out: If your training data is historically skewed — and let’s face it, much of it is — your AI will replicate those inequities. That can mean misidentifying art, erasing Black or Indigenous histories, or telling a story that’s only half-true.
• Be transparent: If you’re using AI in public-facing ways, let your community know how. Who trained the model? What data was used? What safeguards are in place?
• Keep humans in the loop: AI is a tool, not a replacement for curators, archivists, educators, or cultural stewards. Especially when dealing with sacred, painful, or complex histories, you need real people making real choices with real care.
• Don’t skip the privacy bit: If your AI collects data from visitors, make sure your policies are crystal clear — and legally compliant.

Question for the room: Are your AI systems helping repair historical silences — or deepening them?

3. Build Internal Capacity: It’s not just an IT problem — it’s a cultural shift

Too often, AI projects get dropped in the laps of tech teams with little context or collaboration. But this work needs many minds — and a collective willingness to learn.

• Train your team: Not everyone needs to become an AI expert, but your people should understand the basics — what AI can and can’t do, what risks to look out for, and how it fits into their roles.
• Break down silos: The best AI projects bring together curators, educators, IT staff, marketing teams, and even front-of-house folks. Everyone sees something different — and those diverse perspectives are key.
• Lean into partnerships: You don’t have to go it alone. Collaborate with academic institutions, AI researchers, or even grassroots archives doing creative memory work.
• Stay humble, stay curious: AI is evolving fast. What works today might need a re-think tomorrow. Be ready to adapt.

Reflection point: What would it take to build a culture of experimentation, care, and shared learning around AI?

In Closing

The potential for AI in cultural institutions is real — but so are the risks. This tech isn’t just a tool; it’s a mirror, a megaphone, a multiplier. So before we amplify, let’s ask ourselves: what are we choosing to reflect?

If we get this right, AI can be part of a broader movement toward more thoughtful, more inclusive, more dynamic cultural work. But it starts with intention.

So tell me — have you seen AI used beautifully (or badly) in the cultural space? What’s one hope or hesitation you have about the future of AI in memory work?

Drop your thoughts below.