AI analysis reveals £71,000 art as authentic Caravaggio masterpiece
Swiss AI system gives 85.7% probability that painting dismissed by top experts for decades is the part of the 17th-century art work
A painting sold for £71,000 and long dismissed by the art establishment as a mere copy has been identified as an authentic work by the Italian master Caravaggio through a groundbreaking artificial intelligence analysis.
The study of The Lute Player, which belongs to Britain’s Badminton House, concluded with an 85.7% probability that it was painted by the revolutionary 17th-century artist Michelangelo Merisi da Caravaggio.
The finding, described by experts as very high, directly challenges decades of consensus from institutions like Sotheby’s and New York’s Metropolitan Museum of Art, which had labelled the work as inferior.
Dr. Carina Popovici, head of the Swiss firm Art Recognition: “Everything over 80% is very high and a strong match.”
The technology compared the brushwork and artistic style of the Badminton painting to verified Caravaggio works, finding a strong match.
The AI’s verdict turns art market history on its head. In 1969, Sotheby’s sold the painting as a copy for just £750. Even when it last changed hands in 2001 for £71,000, it was catalogued as from the circle of Caravaggio.
The current authentication suggests its value could now be in the realm of tens of millions, given a newly discovered Caravaggio in 2019 was valued at around £96 million.
The painting was purchased in 2001 by British art historian Clovis Whitfield, who has championed its authenticity for years.
-
James Van Der Beek's celebrity pals pen touching tribute after his death
-
James Van Der Beek, 'Dawson's Creek' star, dies at 48
-
Cruz Beckham hints at reunion with estranged brother Brooklyn amid rift?
-
Internet reacts to 10-day flight ban in El Paso
-
Police detained 'innocent' man in Nancy Guthrie's kidnapping
-
Co-founders of Elon Musk's xAI resigned: Here's Why
-
Why Angelina Jolie loves her 'scars' following double mastectomy
-
‘World is in peril’: Anthropic AI safety researcher resigns, warns of global risks