February 2026

Two paintings attributed to the 15th-century Flemish master Jan van Eyck, “Saint Francis of Assisi Receiving the Stigmata,” are facing renewed scrutiny following an analysis utilizing artificial intelligence. The works, housed at the Philadelphia Museum of Art and the Royal Museums of Turin, have long been considered pivotal examples of van Eyck’s artistry, renowned for their realism and innovative oil painting techniques. However, a recent study conducted by the Swiss company Art Recognition, in collaboration with Tilburg University in the Netherlands, casts doubt on their authenticity.

The core of the debate centers on the AI’s inability to detect brushstrokes consistent with those of Jan van Eyck. Art Recognition’s analysis concluded that the Philadelphia version was “91% negative” and the Turin version “86% negative” in terms of exhibiting van Eyck’s characteristic style, raising the possibility that both paintings may be studio works – created by assistants in his workshop rather than the master himself. This development has sparked discussion among art historians and raises questions about the attribution of works by one of the most influential artists in history.

The AI-driven assessment employs a novel method of analyzing paintings to verify authenticity. Art Recognition uses artificial intelligence to identify the unique characteristics of an artist’s technique, and in this case, found a lack of evidence supporting van Eyck’s direct involvement in the creation of these specific pieces. The findings, first reported on February 7, 2026, have prompted a re-evaluation of the paintings’ provenance and artistic merit. The Guardian detailed the initial findings, highlighting the potential implications for the art world.

Jan van Eyck (c.1385-1441) is celebrated for his mastery of oil painting and his ability to create remarkably realistic depictions. Only around 20 works are confidently attributed to him, making each painting a significant cultural artifact. The Philadelphia Museum of Art previously hosted an exhibition, “Recognizing Van Eyck,” in 1998, showcasing several of his works, including versions of “Saint Francis Receiving the Stigmata.” The rarity of van Eyck’s paintings means opportunities to study and exhibit them are infrequent, adding to the significance of this current debate.

Studio Works and Lost Originals

Till-Holger Borchert, director of the Suermondt-Ludwig-Museum in Aachen and a leading van Eyck scholar, supports the idea that the paintings may be studio works. He suggests that the findings align with existing scholarly debates about the extent to which van Eyck personally executed all aspects of his paintings. According to Borchert, the existence of these studio versions could indicate the presence of a lost original painting by van Eyck, one that more fully embodies his unique artistic style. VRT News reported on this perspective, emphasizing the possibility of a more definitive van Eyck version yet to be discovered.

The two versions of “Saint Francis Receiving the Stigmata” were once owned by Anselm Adornes, a Bruges nobleman, further complicating the provenance. The paintings are nearly identical in composition and detail, leading to speculation about their relationship to each other and to any potential original work. The debate isn’t simply about authorship. it touches upon the very nature of artistic creation and the role of workshops in the Renaissance period.

The Role of AI in Art Authentication

Art Recognition’s methodology, as outlined in reports from Paint & Painting, involves analyzing brushstrokes and other stylistic elements to determine an artist’s presence in a work. The company’s approach represents a growing trend in the art world: the application of artificial intelligence to address questions of authenticity and attribution. While AI offers a new tool for art historical analysis, its findings are not without controversy, and are prompting a critical discussion about the limitations and potential biases of algorithmic assessments.

Maximiliaan Martens, a Van Eyck expert, has disputed the AI firm’s conclusions, suggesting the methodology may not fully capture the nuances of van Eyck’s technique. This highlights the ongoing tension between traditional art historical methods and emerging technologies in the field of art authentication. The debate underscores the complexities of attributing artwork, even with the aid of advanced technology.

The implications of these findings extend beyond the two paintings in question. If the AI’s assessment is accurate, it could lead to a broader re-evaluation of works currently attributed to Jan van Eyck, potentially reshaping our understanding of his oeuvre. The Philadelphia Museum of Art and the Royal Museums of Turin have not yet issued official statements regarding the AI analysis, but the art world awaits their response with keen interest.

What comes next will likely involve further scientific analysis and a renewed examination of the historical evidence surrounding these paintings. The debate surrounding the authenticity of these works is expected to continue, prompting further research and discussion among art historians and scholars. Share your thoughts on the role of AI in art authentication in the comments below.