How AI is changing the art world: Art critics fear AI could undervalue their contributions

The dark-haired woman in the picture closely resembles the figure in Pierre-Auguste Renoir’s painting Gabrielle, which was recently estimated at Sotheby’s at $120,000 to Between $180,000.

Art world experts, on the other hand, debate the authenticity of works owned by private Swiss collectors. Artificial intelligence has now stepped in to help settle the debate once and for all, and machine learning has determined that it is most likely the real Renoir.

The Art of AI: Why AI Will Replace Art CriticsPortrait of a Woman (Gabrielle).

Artificial intelligence is increasingly being used to identify genuine works of art and fake priceless treasures. Art Recognition, the Swiss company behind the technology, revealed earlier this month that it had concluded that the only Titian work in Switzerland, the work titled “Night Scene of a Couple” preserved at the Kunsthalle Kunsthaus in Zurich, was very Probably not by 16th century Venetian painters.

However, experts in the art field warn that artificial intelligence (AI) is only as good as the paintings it is trained on. This adds further uncertainty if they are fake or have colored areas.

Art Recognition teaches its algorithm Renoir’s style, characterized by staccato brushstrokes and bold color combinations of figures, using photocopies of 206 real paintings by the French Impressionists. To improve accuracy, it also divides photos into smaller chunks and presents them to the algorithm. It also trained the algorithm on some works by painters with a similar aesthetic to Renoir who were active in the same period. This evaluation concludes that there is an 80.58% probability that Renoir painted Portrait de femme (Gabrielle).

Connoisseurs often tell art owners that their “impressions” or “gut feelings” determine whether a painting is genuine, which can be very frustrating. They greatly appreciate our greater precision.

Dr. Carina Popovici, CEO, Art Recognition

Encouraged by the results, the painting’s owner contacted GP.F. Dauberville & Archives Bernheim-Jeune, a Paris-based organization of specialists, to produce its own catalog of the complete works of Renoir. They also came to the same conclusion after a scientific investigation of the colors in it that the painting was an authentic Renoir.

Dr. Bendor Grosvenor (left)

Dr Bendor Grosvenor, an art historian and presenter of BBC Radio 4’s Lost Masterpieces of Britain, worries that such technology will diminish the importance of expert judgment in determining the validity of artworks.

So far, the techniques used to “train” the AI ​​programs, and their claims that they can determine attribution from just an iPhone photo, haven’t been particularly impressive.

Dr. Bendor Grosvenor

Artificial intelligence is playing a growing role in art appraisal, but it must be combined with provenance research, the knowledge of experts who specialize in artists, and established science such as pigment analysis. It has the advantage that it can answer yes/no questions, such as those involving pattern analysis or matching, and it’s constantly improving, but its output must be interpreted by someone who has to ask the appropriate questions. While we may never fully achieve absolute authentication certainty, we are getting closer.

Learn more about the arts industry:

Information source: Compiled by 0x Information from MPOST.Copyright belongs to the author Damir Yalalov, without permission, may not be reproduced

Related Posts

Crypto whales move over $130 million in XRP and DOGE

Crypto whales moved more than $130 million in Dogecoin (DOGE) and XRP as the crypto markets turned stale this week. An anonymous wallet sent 300 million DOGE to an unknown wallet, totaling over $41.8 million. Earlier this week, an anonymous wallet sent…
Read More