AI Throws a Curveball: Could a “Caravaggio Circle” painting be the Real Deal?
Forget the buzzer-beaters and Hail Marys, sports fans! The art world is buzzing with a potential upset that’s got seasoned experts scratching their heads and a powerful new player – Artificial intelligence – stepping up to the plate.
We’re talking about a painting, one of three known versions of “The Lute Player,” that’s been languishing in the “copy” or “circle of” category for decades. But now, an AI analysis is throwing a massive Hail Mary, suggesting an 85.7% probability that this “Badminton version” is actually a genuine Caravaggio. That’s a stat that would make any stat-head do a double-take!
For years, this particular “Lute Player,” originally from Badminton house in England, has been treated as a lesser imitation. Sotheby’s, in 1969, even listed it as a mere copy “after Caravaggio,” fetching a mere 750 pounds. Fast forward to 2001, and it was bumped up slightly to the “Caravaggio circle” for around 71,000 pounds. That’s a decent jump, but still a far cry from the big leagues.
Meanwhile, other versions of “The Lute Player” are in the hallowed halls of the Hermitage Museum in St. Petersburg and the prestigious Wildenstein collection, which even saw time on loan at the Metropolitan museum in New York.
But here’s where the plot thickens, like a perfectly executed play. British art historian clovis Whitfield, who co-owns the Badminton version with the late collector Alfred Bader, is calling a flag on the play.He’s convinced the mainstream experts are missing a crucial detail, much like a referee overlooking a blatant foul.
Whitfield points to a 1642 biography of Caravaggio by Giovanni Baglione.He argues that Baglione’s description of a painting with this subject matter “corresponds exactly” to the Badminton version. Baglione mentally observed details, such as the reflection of the drops of dew on the flowers,
Whitfield asserts, suggesting the original biographer was describing this very piece.
Though, not everyone is ready to put this painting in the Hall of Fame just yet. American art historian Keith Christiansen, who previously headed the European painting section at the Met, has been a vocal skeptic. Back in 2007, he reportedly told Whitfield that Nobody, and certainly no modern scholar, has ever thoght or would ever think that his painting may have been made by Caravaggio.
Christiansen declined to comment for this report, but his past stance highlights the deep-seated opinions within the art establishment.
Enter our new MVP: Artificial Intelligence. Carina Popovici, founder and CEO of Art Recognition, a Swiss company pioneering AI in art authentication, revealed the AI’s findings. The technology found a strong correspondence
between the Badminton “lute Player” and a dataset of known Caravaggio works. Popovici emphasizes that Any certainty greater than 80% is very high,
making that 85.7% a game-changer.
Popovici believes this situation perfectly illustrates the rigidity of the current market and the way in which the opinions of the highly rooted experts can hinder progress.
She hopes that AI, as a non-human tool, can help face these situations with greater opening and less conflicts.
What does this mean for us sports fans?
Think of it like a controversial replay call. For years, the refs (art historians) have made their call. But now, with advanced technology (AI), we’re getting a new viewpoint. Could this be the equivalent of a VAR review overturning a bad call?
This debate raises captivating questions for U.S. sports enthusiasts:
* The “underdog” Story: We love a good underdog story in sports. Could this painting be the ultimate art world underdog, finally getting its due thanks to a technological assist?
* Data vs. Gut Feeling: In sports analytics, we’ve seen data increasingly inform decisions, sometimes challenging conventional scouting or coaching instincts. Is AI the “Moneyball” of the art world?
* The Future of Authentication: If AI can accurately assess art, what does that mean for the future of sports memorabilia authentication? Could we see AI analyzing game-worn jerseys or signed baseballs with similar statistical rigor?
Potential Areas for Further Investigation:
* AI’s “training Data”: What specific Caravaggio works was the AI trained on? Understanding the dataset is crucial to evaluating the AI’s conclusions.
* Human Expert Collaboration: How can AI and human experts collaborate more effectively? Rather of an adversarial relationship, can they work together like a coach and a data analyst?
Keep reading
