Harvard Art Museums has explored a fresh way to connect audiences with art by using generative AI in a marketing pilot centered on an 18th-century painting from its collection. The experiment aimed to boost visitor engagement and membership while making the museum experience feel more interactive.
The project was developed with input from John Connolly, the museum's marketing director, and Julian De Freitas, an assistant professor at Harvard Business School. Their case study transformed François Boucher's 1750 portrait Jeanne-Antoinette Poisson, Marquise de Pompadour into a talking video version and paired it with a chatbot that could answer questions in conversation form.
According to the team, the idea was not simply to use AI for novelty, but to test whether technology could help people feel closer to historic art. The pilot was shown to selected participants in different versions, offering a glimpse into how audiences respond when classic works are presented through a digital lens.
Feedback was mixed, yet many participants responded positively to the concept and the technology behind it. The museum team noted that reactions often reflected broader public attitudes toward AI, while others appreciated the humor and the unexpected format.
De Freitas said the project highlighted the need for sensitivity when introducing AI into cultural spaces. He emphasized that the technology must add genuine value, while Connolly described the effort as part of a wider mission to make museums feel more open, playful, and approachable.
With its vast collection and free admission model, Harvard Art Museums continues to rethink how institutions can attract new audiences in a changing digital era. The experiment suggests that AI may become a creative bridge between heritage and modern engagement, shaping how museums connect with future visitors.
