Celebrating 15!

Exploring GenAI on site at The Met

How we developed our Google-supported technologist residency.

Interior View of the Metropolitan Museum of Art when in Fourteenth Street (1881) by Frank WallerThe Metropolitan Museum of Art

Evolving over time

Over our 156 year history, The Met has witnessed—and been a part of—many radical shifts in the way visitors experience art.

Ancient Rome Ancient Rome (1757/1757) by Giovanni Paolo PaniniThe Metropolitan Museum of Art

Exploring parallel futures

Each technological leap offers many paths forward, often offering a huge array of possibilities, ideas, and challenges.
Generative AI is one such leap, and, since it's a very broad and transformative technology, the number and scale of possibilities is dizzying.

Our idea: explore this complexity head-on through a new, focused position, a "Technologist in Residence." We decided to start with the on-site experience, and set out to develop as many experiments as possible: what’s working? What’s not? What do our visitors want?

User-Testing the Unlikely Pairings Prototype by The Metropolitan Museum of ArtThe Metropolitan Museum of Art

6 Months, 3 Sprints, 20+ Prototypes

We embedded a creative technologist within the Met for six months to explore how AI might impact the visitor experience at the museum. They rapidly built prototypes using Gemini and Vertex AI, and tested them on the floor with visitors, exploring both what works and what our visitors want and need.

The Met's Technologist in Residence - Structure by The Metropolitan Museum of ArtThe Metropolitan Museum of Art

The sprints

We broke up the 6 months into 3 focus areas:
"Wayfinding," helping visitors look broadly across the museum to find what's relevant and of interest

"Close Looking," diving deep into an individual work or gallery

 "The Met and the World," connecting to data and experiences outside our walls.

The Met's Technologist in Residence - Process by The Metropolitan Museum of ArtThe Metropolitan Museum of Art

GenAI supercharges iteration

AI coding agents made it feasible to workshop a concept Monday, build a working prototype Tuesday/Wednesday, test on Thursday, and document on Friday. 

This, on top of a defined process and support structure, made this pace and process possible.

Our experiments led to a broad array of insights. One above all stood out.

User-Testing the Wall Label Prototype by The Metropolitan Museum of ArtThe Metropolitan Museum of Art

Transformative AI wins the day

Our visitors were excited about projects that took full advantage of AI: offering content in dozens of languages, adding opportunities for deep personalization, etc. They were skeptical of ideas that don't meet that threshold.

User-Testing AI Prototypes by The Metropolitan Museum of ArtThe Metropolitan Museum of Art

Dive in

In the following three posts, our resident Julia Daser will dive deep into each sprint: what worked, what didn’t, and what we learned from our visitors. Enjoy.

Credits: All media
The story featured may in some cases have been created by an independent third party and may not always represent the views of the institutions, listed below, who have supplied the content.
Explore more
Related theme
The Met: Prototypes & Play
How the museum is using AI to explore its collection in new ways
View theme

Interested in Performance?

Get updates with your personalized Culture Weekly

You are all set!

Your first Culture Weekly will arrive this week.

Home
Discover
Play
Nearby
Favorites