Part 2 of a Three-Part Series on Human-Centric AI for Cultural Institutions
In Part 1, we explored how museums are carefully — but enthusiastically — embracing AI to support their missions. Now, as more institutions look to put these tools into practice, the focus shifts to doing it thoughtfully and well. This article highlights six clear principles for making AI adoption a positive force in your museum.
If the first article in this series made one thing clear, it’s this: museums are recognizing AI as a powerful tool for progress. Rather than holding back, cultural organizations around the world are actively exploring how to use AI to enhance their work and engage their communities.
The big question now is: If we’re going to use AI, how do we make the most of it?
The best answer so far is a “human-centric” approach, a set of principles that ensure technology supports people, not just algorithms or trends. Drawing on research and direct interviews with museum professionals, Cuseum has built these sets of recommendations that, with the help of certain digital tools, can turn AI into a practical, positive change.
1. Visitor-Centered Design: Start With People, Not the Tech
Every digital tool, from a simple chatbot to a sophisticated recommendation engine, should be built to make visitors’ lives easier. That means more than just convenience. You should focus on providing accessibility, inclusion, and real engagement.
AI-powered translation and personalized tours are already helping break down language barriers and make collections more approachable. As one practitioner put it:
“AI’s capabilities are central to enhancing content accessibility, especially for international audiences.”
Example:
The Rijksmuseum’s AI-driven image retrieval system lets visitors explore artworks through color and theme, opening up new, personalized ways of engaging with collections.
Takeaway:
AI should meet visitors where they are; on mobile, in their language, with easy navigation and clear information.
2. Preserve the Curatorial Vision and Institutional Identity
Museums are trusted for their expertise and judgment. AI should amplify those qualities. The approach implemented, or the tools that are enhanced with AI, can help analyze visitor data, but curators ultimately make the final decisions.
As one anonymous curator talking with Derda and Predescu stressed: “Museums should not give up their unique position as spaces providing reliable information. Technology is here to help us preserve our spot as educational channels and access even more people at that.”
How this plays out:
Use AI for routine tasks: tagging collections, generating drafts, crunching numbers.
Maintain a clear human review and approval process for anything public-facing or interpretive.
Make sure AI tools are supporting, not replacing, your institution’s authentic voice.
3. Ethical AI Governance: Build Trust by Design
No one wants a scandal about visitor data or algorithmic bias making headlines. Ethics have to be built in from the start, especially in public institutions with an educational mission.
Key practices:
Transparent data policies. Tell visitors what’s collected and why.
Regular audits of AI tools for bias and accuracy.
Limit use of personal data to only what’s necessary for a better visitor experience.
As the research puts it: Clear ethical guidelines must govern the implementation of AI, addressing concerns around data privacy, algorithmic bias, and transparency. Ethical practices safeguard visitor trust.
4. Support and Complement Human Efforts—Don’t Replace Them
Automation can save time and resources, but museums will always be, fundamentally, about people. AI should take on repetitive tasks and free up staff to do more meaningful, creative, and visitor-facing work.
AI should enhance, not replace, human capabilities within museums. By automating routine tasks, AI can free up staff to focus on creative, educational, and interactive roles.
Practical impacts:
Use chatbots for after-hours questions, so staff can focus on deeper engagement during open hours.
Let machine learning handle visitor flow analytics, while your team concentrates on cultural events, audience engagement, and hospitality.
5. Human Oversight and Control—Keep Humans in the Loop
Every AI tool in a museum should have clear boundaries, with humans able to intervene, review, and override.
This is not just a technical issue, but a cultural one. It’s about safeguarding the museum’s mission and maintaining public accountability.
Maintaining human control over AI systems is essential to ensure accountability and reliability. AI tools must be set up to remain aligned with institutional values and priorities.
6. Use Museums as Platforms for AI Discourse
Museums are not your average user of technology, they’re also places for critical dialogue about it. The best institutions use their position to help visitors understand AI’s societal impacts and potentials.
Real-world inspiration:
A growing number of institutions now offer digital literacy workshops, AI-themed exhibitions, and public forums to explore the ethical and creative sides of emerging tech.
🏛️ Museums Using or Working around A
These are just some of thousands of institutions that have already incorporated AI into their organization or have built exhibits where AI acts as a center piece.
1. Museum of African American History (Boston, USA)
This museum launched “Black Voices of the Revolution,” an exhibit where AI-powered digital narrators bring personal and lesser-known stories from Black history to life.2. Palace of Versailles (France)
At Versailles, AI chatbots are embedded in twelve garden statues, allowing people to have real-time conversations with historical figures from the palace’s past.3. National WWII Museum (New Orleans, USA)
In “Voices From the Front,” AI avatars of WWII veterans answer guests’ questions with unscripted, story-rich responses drawn from real oral histories.4. Imperial War Museums (United Kingdom)
The museum made its immense collection of wartime audio archives accessible with AI transcription, turning thousands of hours of recordings into a searchable resource.5. MIT Museum (“AI: Mind the Gap”, Cambridge, USA)
Here, the “AI: Mind the Gap” exhibition explores the evolution, ethics, and influence of artificial intelligence through interactive displays and thought-provoking installations.6. Seoul Robot & AI Museum (Seoul, South Korea)
Breaking new ground, this institution was designed and partially built by robots. Its galleries offer hands-on encounters with the latest in AI and robotics.7. Art Center Nabi (Seoul, South Korea)
Since 2016, Art Center Nabi has hosted pioneering exhibitions where artists collaborate with AI, making the machine a true creative partner on the digital canvas.
Turning Principles Into Practice: The Human-Centric AI Governance Model (HC-AIM) Model
These six principles are the backbone of the Human-Centric AI Governance Model (HC-AIM) designed by Izabela Derda and Domnica Predescu in 2025, and first implemented in Europe but now applied in multiple institutions around the world. It’s a flexible framework any museum can adapt, regardless of size or technical capacity.
What it looks like in action:
Start small with pilot projects. Test, gather feedback, iterate.
Involve staff from all levels in decision making, not just IT or digital.
Communicate clearly with your public about what’s new, why it matters, and how it benefits them.
Make ethics, inclusion, and accessibility part of every technology conversation.
SIDEBAR: Human-Centric AI in Museums – Do’s & Don’ts
DO:
Design for Visitors First:
Make every AI feature about better access, inclusion, and engagement for all audiences.Prioritize Transparency:
Tell visitors how their data is used, what AI does, and why.Keep Curators in Charge:
Use AI to support, not replace, human expertise and institutional voice.Audit Regularly for Bias:
Test and review for unintended bias or skewed results.Make Ethics the Default:
Build in privacy protections and give visitors control over their data.Pilot and Iterate:
Start with small projects, gather feedback, and improve continuously.Use Your Platform:
Offer programs, talks, or exhibitions about AI and its impact on society.
DON’T:
Don’t Automate Everything:
Resist the urge to replace meaningful human roles with automation.Don’t Ignore Accessibility:
Don’t let AI tools create new barriers or distractions for visitors.Don’t Treat AI as Plug-and-Play:
Avoid “set it and forget it.” AI in museums requires ongoing oversight.Don’t Hide Behind the Algorithm:
If something goes wrong, don’t blame the tech—be accountable and transparent.Don’t Collect Data Without Purpose:
Only gather what’s necessary, and explain the benefit to your visitors.Don’t Forget the Mission:
Technology should serve your educational and cultural goals—not the other way around.
Want a quick diagnostic?
Ask yourself the following questions:
“Does this tool make our visitors’ experience better?”
“Can we explain how it works and why we use it?”
“Who makes the final call, the algorithm or a human being?”
If you can’t answer with confidence, it’s time to revisit your approach.
Next Up:
In the final part of this series, we’ll look at real museum case studies, where AI tools are working well, what to avoid, and how to create your own roadmap for responsible digital innovation.