NYC Gazette

In recent years, artificial intelligence (AI) has begun to play a significant role in art conservation, a field steeped in history and science. This development is being explored by New York University's Institute of Fine Arts (IFA), led by Director Joan Kee. According to Kee, "This is a negotiation between the machine gaze and the human gaze. The potential is great, because there are many who are trying to think more systematically about their analysis."

IFA recently hosted a panel discussion in February, bringing together specialists in art history, conservation, and computer science to discuss the implications of AI in their fields. Additionally, in collaboration with NYU's Tandon School of Engineering, IFA is offering its first course on AI conservation. Kee notes, “AI can provide valuable insights in the analysis of art, but it should be viewed as a tool rather than a replacement for human expertise."

AI technologies are being used to assess authenticity, restore damaged artworks, and support provenance research. According to Kee, "AI is being leveraged to analyze patterns in materials and artistic techniques," enhancing traditional conservation methods. A notable example cited by Professor Christine Frohnert involves AI's role in decoding the carbonized papyrus scrolls from Herculaneum, revealing hidden texts that were previously unreadable.

While the potential benefits are significant, Kee acknowledges that AI's effectiveness is contingent on the available data. "Conservation is a small field and it produces relatively limited datasets," she explains. Artworks from well-funded museums have more comprehensive data, unlike less documented artifacts and non-Western works. Similarly, Professor Frohnert highlights issues with data standardization, which can affect the reliability of AI analysis.

The course "Conservation of AI-Based Artworks" is being co-taught by Professors Deena Engel and Thiago Hersan. It aims to merge the skills of art conservation, art history, computer science, and engineering to foster innovation in media conservation. By examining works by Rafael Lozano-Hemmer, students are exposed to the programming languages crucial for conservation and develop strategies to preserve these artworks for the future. The course is part of a broader effort to establish a new Conservation Technologies program at NYU.

Kee’s vision is for AI to complement, not replace, traditional conservation methods, as the field continues to evolve with technological advances. The path forward for AI in art conservation includes overcoming challenges such as limited data diversity and standardization issues while maximizing its potential to assist human expertise.