AI Won't Tell You What Art Is Worth—But It Might Decide What You See
Artificial intelligence is rapidly becoming part of the art world. From personalised recommendations on digital platforms to image-recognition tools used by auction houses, AI is changing how artworks are discovered, compared, and presented to collectors. Yet one question continues to dominate the conversation: Can AI value art?
The answer is more complicated than a simple yes or no.
While AI can process enormous amounts of visual and behavioural data far faster than any human, it cannot determine artistic value in the way many imagine. Unlike financial markets, where value can often be measured through numerical indicators, the art market depends on history, reputation, cultural significance, provenance, exhibitions, critical reception, and institutional recognition. These are qualities that algorithms alone cannot fully interpret.
Instead of replacing experts, AI is increasingly acting as a powerful assistant.
Today, platforms such as Artera personalise artwork recommendations by analysing visual similarities and user behaviour, while systems like Sotheby's former Thread Genius technology help specialists identify comparable works more efficiently. Artsy's Art Genome Project organises thousands of artworks through detailed characteristics—or "genes"—making discovery easier for collectors and researchers alike. Together, these technologies reshape how people encounter art long before a purchase is ever made.
Perhaps the greatest contribution of AI is not valuation itself, but visibility.
Algorithms influence which artists appear in front of collectors, which artworks are recommended, and which connections become visible across vast collections. In an increasingly digital market, visibility matters. The works we discover often become the works we research, discuss, and eventually collect.
Yet visibility should never be mistaken for value.
An algorithm may recognise that two paintings share similar colours, compositions, or subjects, but it cannot understand why one transformed the history of art while another did not. It cannot assess the significance of an artist's career, the influence of a particular exhibition, or the cultural conversations surrounding a work. Those judgments remain rooted in human expertise.
This suggests that the future of AI in the art market lies in collaboration rather than replacement.
Rather than acting as an autonomous appraiser, AI functions as an intelligent filtering system. It identifies patterns, surfaces potential comparables, and personalises discovery, while curators, specialists, galleries, and collectors provide the historical context and critical interpretation that transform information into meaning.
As AI continues to evolve, the challenge for the art world will not be deciding whether to adopt these technologies, but learning how to use them responsibly. Questions surrounding transparency, algorithmic bias, and institutional authority will become increasingly important as digital tools shape what—and who—receives attention.
The future of art valuation, therefore, is unlikely to belong entirely to humans or machines. It will belong to the conversation between them—where AI organises information, and human expertise continues to create meaning.