Artificial intelligence (AI) is worrying due to its ability to imitate art, although there are new companies specialized in this technology, such as Art Recognition, which show that it can also help in the fight against artistic forgeries, according to its manager, Carina Popovici.
The firm, a pioneer in the use of AI to authenticate works of art, recently attracted attention after discovering that more than 40 counterfeit paintings, including an alleged work by Monet and another by Renoir, were trying to be sold on eBay.
You may be interested in: Robots: From domestic assistants to planet explorers
“Despite the perception that AI poses a threat to art connoisseurs, that is far from the truth, and is instead dedicated to complementing art experts and traditional authentication methods,” he noted. Popovici, CEO and founder of the company.
Popovici created this company in 2019, based in a development center for emerging brands in the small town of Adliswil, on the outskirts of Zurich, and already works with art galleries, auction houses and collectors.
How does it work
“We use an AI based on a convolutional neural network, which can learn the main characteristics of an artist from authentic photographs and subsequently recognize them in a new work of art not previously seen,” explains the software designer.
To do this, the Artificial Intelligence must first go through a training phase, in which it examines for a period of between one and three days photographs of all the artist’s known works and negative examples of forgeries if any, using service providers. comprehensive cloud data center solutions, such as AWS and Azure.
One of the main characteristics that helps the software verify whether or not the work is the author’s is the brushstrokes used on the canvas, but the AI also analyzes other elements such as the color palette used and the level of composition.
“Traditionally, art authentication has depended on the experience and judgment of experts, who are still the dominant authority in the sector, but this approach is subjective, and opens the possibility of human error,” he points out.
Art Recognition is not presented as infallible either (it presents its results with a probability percentage that in recent cases on eBay exceeded 90 percent) and according to the head of the firm it also has its limitations.
Impressionists, the easiest to verify
This technology cannot currently verify a Vermeer work, given that only about 36 exist (the larger the database, the more precise it is) and it cannot effectively distinguish highly restored works, such as the ‘Salvator Mundi’ that is still preserved. discusses whether or not Leonardo da Vinci painted.
In contemporary art, this new tool has problems verifying copies of the famous drippings (works based on splashes and jets) by Jackson Pollock or the originals by Modigliani, a creator about whom experts still do not agree on what are his true works.
In any case, Popovici offers his company’s work to “contribute to a more transparent art market.”
A sector that, he acknowledges, is threatened by AI itself, “capable of replicating the styles of renowned artists, producing fakes that flood the market” but which, like so many advances of the past, is a double-edged tool that at the same time can benefit art.
“We train our algorithm to differentiate authentic pieces from AI-generated fakes, also feeding the program with digital fakes,” he says.
He adds that this technology can not only help in the fight against imitations but also already participates in the art world through interactive experiences in museums and galleries, or in data analysis of market trends.
“In addition, generative AI models are revolutionizing the way artists explore creativity and I am sure we will see even more spectacular AI-assisted works in the future,” she argues.
➡️Join the El Sol de México channel on WhatsApp so you don’t miss the most important information
Art Recognition has been tested against works in which there are extensive debates about their authorship, such as a “Samson and Delilah” exhibited in the National Gallery in London and attributed by some experts to Rubens.
The software designed by her considered that the work was not by the Flemish master, while in a similar test with a self-portrait of Van Gogh treasured by the National Museum of Oslo it did declare its authenticity.
2024-05-20 11:00:00
#Artificial #Intelligence #art #forgeries #Sol #México
