AI reveals that Raphael may not have painted this entire iconic work of art

AI may have cracked the code to a long-standing question about a famous Raphael painting that has puzzled the art world for years.

Part of Raphael’s Madonna della Rosa may not have been painted by the Italian Renaissance master, according to findings published in the Science Heritage Journal in December 2023, which have now resurfaced and are circulating online.

Raffaello Sanzio da Urbino is considered one of the most famous artists of the Renaissance era for his paintings and frescoes, including the “School of Athens” in the Vatican.

But the Madonna della Rosa, which is on display at the Museo Nacional del Prado in Madrid, has long sparked speculation about its attribution, with the depiction of St John appearing out of step with Raphael’s style and less refined than other elements of the painting. The picture shows the Madonna and baby Jesus together with St. Joseph and St. John.

Computer scientists from the universities of Nottingham, Bradford and Stanford developed a specific AI algorithm to compare parts of the painting with other works by Raphael.

The team photos used from verified Raphael paintings to “train the computer to recognize his style to a very detailed degree, from the strokes, the color palette and the shading and every aspect of the work,” according to Hassan Ugail, a computer scientist at the University of Bradford.

“When we tested della Rosa as a whole, the results were not conclusive,” Ugail said in a statement. “So we then tested the individual pieces and while the rest of the photo was confirmed as Raphael, Joseph’s face came out as most likely not Raphael.”

In a press release, computer scientists said it is believed that one of Raphael’s students, Giulio Romano, may have painted the face of St. John. Although the findings are not 100 percent confirmed, Ugail noted that the computer model examined artwork at the microscopic level with 98 percent accuracy.

Christopher Brooke, an honorary research fellow at the University of Nottingham who is also involved in the project, said in the press release that the analysis “promises to be a useful, complementary tool in future studies of this nature alongside established methods such as spectroscopy.”