Unlock the editor’s summary for free
Roula Khalaf, editor of the FT, selects her favorite stories in this weekly newsletter.
When I worked at Christie’s as a student of a summer, the book of the book collector William Foyle went on sale. There were about a thousand lots sold for three days. Only one book was retained, he could never on sale. He had a Nazi seal inside him.
The increase in origin of origin, the documented history of an object, is directly connected to forced sales, appropriations and robberies during World War II. So much cultural heritage changed from illegally that there is a significant danger that a previous owner intervenes with a sale. It can get messy, both morally and legally.
Art is authentic in three main ways. The oldest is the connoisseur: experts (often self -proclaimed as such) provide their opinion on who created a work of art. The research of origin is the second, which means gathering historical documents and references to the object in question, which attests generations of experts who agree on their authenticity and a document trace that affirms its legal property. The third are forensic tests.
Artistic trade has always used forensic tests as a last resort, a tiebreaker when experts disagree or the origin raises a red flag. There is no good reason for this. Forensic tests do not need to be expensive or invasive: it often implies looking an object with different spectra of light (radiographs, infrared, ultraviolet). But artistic trade has always worked in the agreements of the Knights and the opinion of the connoisseurs. A forensic test would confirm what the expert already said (therefore, not necessarily add value from a commercial perspective) or contradict the expert (in which case the expert loses and art trade loses a possible sale).
In the history of falsification, tricks have often created “origin traps” to take advantage of excessive dependence on origin in origin. The basic premise is that the counterfeit passes its falsification as a lost work of a renowned artist, and an enthusiastic expert, who feels that they are in a real -life treasure search. Almost none of the falsifications that I have studied would resist basic forensic evidence. They did not have to do it, because an attractive origin made an object seem good enough to sell.
But we have entered a new era of distrust. Populism distrust experts. Experts distrust the data because, in the era of AI, photographs and texts and even videos can be forged so perfectly that it can be almost impossible to distinguish the facts of fiction. The origin, which deals mainly of Mohoso archive documents, feels more tangible than the opinion and less responsible for being manipulated than the scientific evidence that most people do not understand. So where to go from here?
Technology can manipulate and deceive, but it can also be used forever. Consider outfits such as the C2RMF headquarters in Paris that uses the latest forensic tools to examine pigments and analyze works with varied light spectra. British museum researchers are currently using advanced techniques such as computerized tomography, molecular and isotopic analysis, and radiocarbon dating in a study of mummies of Egyptian animals, revealing the methods used by ancient reservoirs. There are also new initiatives, such as the artistic recognition based in Zurich (which I have joined as an advisor), which uses AI models to authenticate works of art, as well as to evaluate where several artists may have been involved (for example, through hard -hand restorations that can confuse the account of the account in the account).
At this year’s art business conference in Tefaf in Maastricht, the founder of Artistic Recognition, Carina Popovici, will talk about a recent case. They tested a “Bath of Diana” paint from a private collection, through a curing set that included 329 images of confirmed Rubens works, as well as 316 images of non -authentic works used to teach the AI what No be deceived by. The result confirmed that the painting is partially by Rubens (and specified what parts).
While the chemical analysis and the images of spectra of light are good to detect anachronisms (a pigment that after the date of the supposed creation of a paint, or the lack of fingers on a canvas that should have them), these tools are not so good to identify the authorship. IA art analysis isWhenever you have a closed and well -curited data set stored with images by the alleged artist.
Forgivers have always understood that they do not have to make such bright falsifications, provided that their trace of origin is convincing enough. If all works of art for sale were expected to have forensic results that accompanied them, together with the opinion of experts and the documentation of origin, then we would see much less attempts. It has never been so easy to analyze art scientifically. These mandatory forensic evidence would bring a very necessary tranquility to the market.
Discover first about our last stories – FOLD FIGHT FT in Instagram and unknownand register To receive the FT weekend bulletin every Saturday morning