The Center Pompidou is using artificial intelligence to improve visitor experience
Conversational AI offers an opportunity for museums and galleries to serve their visitors in a new way. The Center Pompidou is a great example of how to use technology to make the visitor experience easier and more engaging. They create an AI chatbot with the help of Ask Mona to provide visitors with a non-audio-guide alternative to traditional gallery tours.
At the MuseumNext Digital Summit 2022, Augustin Pagenot and Alix Deltour shared their thoughts on the future potential of Conversational AI in the museum space.
What is conversational AI
Conversational AI has the ability to understand freeform text or speech from humans. Instead of offering a menu of preset options, people can speak to the AI chatbot naturally and ask questions. The AI chatbot interprets the intention behind the question and provides a helpful response.
Conversational AI can do this because it has the ability to learn. Before a chatbot is released to the public, it has to be “trained.” Training consists of feeding it as much data as possible on the ways people may ask for certain types of information. The AI technology learns from context what information to provide based upon intention cues.
Conversational AI offers a more interactive experience to gallery visitors and gives them the ability to quickly find the information they are looking for.
Uses of conversational AI in a gallery space
The Centre Pompidou use their chatbot to perform two main functions:
- An information centre – visitors can ask the chatbot questions to prepare for their visit or while they are in the gallery to make the most of their experience. They can find information about what’s on at the gallery, details about the collection, or even get help finding amenities.
- A virtual tour guide – by snapping photos of their favourite artworks, visitors can learn more information, just like having their own tour guide. The chatbot can recognise the artwork and tell them information about the piece.
The first function can be used while visitors are visiting the Centre Pompidou or beforehand so they can plan their visit. One challenge the team faced in implementing this feature was teaching the conversational AI how to interpret all of the questions it may receive. The Centre Pompidou has a large schedule of events that cater to many different types of visitors.
The beauty of conversational AI is that it is a living technology. You can update it on a daily basis and give it corrections to ensure it is serving your visitors in a helpful manner. The more data it receives, the more effective conversational AI chatbots will be. Gallery staff can feed it additional data and review its suggestions, and it also gets data when visitors use it.
Currently, the Centre Pompidou chatbot serves English and French-speaking visitors, but they plan to expand to cover Spanish-speaking visitors too.
Challenges of conversational AI for art galleries
The main challenge Centre Pompidou found in using conversational AI was choosing the “face” of the art gallery. The conversational AI would essentially be a representative of the Centre Pompidou, so it was important that it communicated with visitors in a way that fit the gallery’s brand. The chatbot’s voice, tone, and effectiveness would all reflect on the Centre Pompidou.
According to Augustin Pagenot from the Centre Pompidou, it took around 8 meetings just to conceptualise this chatbot avatar.
MuseumNext offer online learning for museum professionals striving for engaging, relevant and flexible professional growth content. Learn more about our virtual museum conferences here.