France’s Transport Minister, Philippe Tabarot, says the country won’t authorize Tesla’s “Full Self-Driving” (Supervised) “as it currently stands,” pointing to two safety problems: the system lets drivers speed, and it doesn’t keep them attentive enough in cities.
But this isn’t France slamming the door on Tesla. It’s France refusing to rubber-stamp the Netherlands’ national approval and holding out for a proper EU-wide sign-off.
France and Germany want to quit relying on America and China for key technology like artificial intelligence, but they’re having to choose where to do it.
Ce n’est pas une canicule comme les autres. Ce 23 juin 2026, la France a basculé dans un autre climat. Records pulvérisés, températures dignes du Sahara, chaleur inédite à l’échelle nationale : en une seule journée, le pays a vécu ce que les scientifiques redoutent depuis des décennies. Et ce...
Des scientifiques chinois montrent que l'ensoleillement sur Terre est en train d'être modifié, et d'être redistribué de manière différente à travers la Planète. Certaines régions du monde seront de plus en plus lumineuses, tandis que d'autres vont s'obscurcir, et le phénomène semble avoir déjà...
Le 12 août prochain, une éclipse solaire offrira un spectacle rare aux observateurs du monde entier. Visible depuis la France métropolitaine, elle promet des conditions d’observation particulièrement impressionnantes dans certaines régions. Mais attention : admirer le Soleil, même partiellement...
Pour sa dixième édition, VivaTech 2026 révèle une France de l’innovation moins parisienne que d’habitude. De la santé numérique à l’hydrogène, du quantique à la mobilité décarbonée, les pavillons régionaux veulent prouver que les prochains champions de la tech se trouvent aussi à Rouen,...
La Coupe du Monde 2026 s’ouvre sous tension pour les Bleus. Ils sont prêts à courir, à encaisser des chocs… mais leur adversaire le plus redoutable ne sera peut-être pas celui qu’on croit. Chaleur, humidité, altitude : selon les scientifiques, ces paramètres invisibles pourraient peser lourd sur...
Researchers from the Université Polytechnique Hauts-de-France have revealed a technique that successfully analyzed several Vincent van Gogh artworks at a microscopic level to correctly identify well-known, previously identified forgeries from recently authenticated ones
The research team behind the pioneering, non-invasive art authentication process said their approach could provide museums, collectors, and art auction houses with a previously unavailable tool with a proven, scientifically based track record for identifying fake or counterfeit art.
Using Microscopic-Level Surface Scans to Foil Art Forgery
According to a press release announcing the successful counterfeit identification tests, art forgery is an increasingly complex problem. However, the traditional approach of expert opinion combined with historical research and invasive pigment analysis has not always proven successful, with the study authors noting that such methods are “often inconclusive,” leaving museums and collectors to shoulder most of the risk for difficult-to-authenticate works.
More recently, collectors and auctioneers have turned to science for answers, including a report showing that AI may have identified as many as 40 counterfeit artworks sold on online marketplaces. Although no counterfeit versions of Van Gogh’s works were found, that same effort identified potentially fake versions of paintings by Monet and Renoir that were indistinguishable from authentic ones.
In 2023, researchers at the University of Sydney announced the development of a cutting-edge microscope capable of surpassing the diffraction limit. In that story, study co-author Professor Boris Kuhlmey noted that the method, which is ideal for identifying microchip irregularities, “could even be used to reveal hidden layers in artwork, perhaps proving useful in uncovering art forgery or hidden works.”
Last month, The Debrief reported the discovery of a mathematical pattern in many famous expressionist artworks, noting that many abstract artists may naturally arrange shapes and patterns in similar ways, even without knowing the mathematics behind them. If correct, this concept could use science and mathematics to separate fake Impressionist works.
Claude Monet painted Impression, Sunrise (Impression, soleil) in 1872, unknowingly giving a name to an art style and movement that science says contains an underlying mathematical pattern. Image credit: Musée Marmottan Monet, Paris.
Now, the Université Polytechnique Hauts-de-France team says their new approach takes modern technology a step further, resulting in an unprecedented level of accuracy.
“Fractal Analysis Gives Us a Measurable Fingerprint of an Artist’s Brushwork”
Unlike the above techniques, the new approach analyzes the texture of a painting’s surface at the microscopic level by converting high-resolution images into three-dimensional surface ‘maps’ of the painting’s upper layers. According to the study authors, the extracted “texture” of a painting allows them to measure an image’s roughness and detail using fractal math. For example, the team said that the subtle patterns of an artist’s brush stroke are so consistent “that they act like a morphological signature unique to that artist.”
“Fractal analysis gives us a measurable fingerprint of an artist’s brushwork without needing to sample or disturb the painting,” explained Francois Berkmans, the study’s lead researcher.
To test their approach, the researchers scanned several artworks attributed to Renaissance artist Vincent van Gogh. The researchers also used their technique to study the mathematical texture of an artwork called The Plowmen, which the team noted is a “well-documented fake.” Last, they used their method to analyze another Vincent van Gogh work, Sunset at Montmajur, which was authenticated in 2013.
Sunset at Montmajour is a landscape in oils painted by the Dutch artist Vincent van Gogh on 4 July 1888, but was not authenticated until 2013. By Vincent van Gogh – 1. twitter.com2. nrc.nl, image, Public Domain, https://commons.wikimedia.org/w/index.php?curid=28169507
After running their microscopic surface analysis, the team’s new system identified The Plowmen as a “strong outlier” compared to the artist’s other works. Conversely, the new method determined that the two newly authenticated paintings “aligned closely with Van Gogh’s known works.”
New Method “Strengthens” Traditional Analysis as a Complimentary Tool
When discussing the study’s results, Berkmans said their technique “can clearly point out genuine artists and reliably detect known forgeries.” Along with identifying fakes and authentic works, the team said that the new approach successfully separated Van Gogh’s stylistic signatures from those of 17th‑century painter David Klöcker Ehrenstrahl, an achievement they heralded as demonstrating the technology’s “wider potential.”
Although the new approach had successes, the team said it is not designed to replace current approaches, including complementary analyses such as the chemical examination of materials. Instead, they said their surface-texture approach should be used as a complementary tool to strengthen authentication, resulting in a multifaceted approach that can reduce risk for collectors, museums, and auction houses alike.
“This approach won’t replace traditional expertise, but it significantly strengthens it.”
Christopher Plain is a Science Fiction and Fantasy novelist and Head Science Writer at The Debrief. Follow and connect with him on X, learn about his books at plainfiction.com, or email him directly at christopher@thedebrief.org.
Deux astronautes français dans l'espace : l'un à bord de l'ISS, l'autre sur la future station privée américaine Haven-1. L'annonce, faite par Emmanuel Macron lors de « Choose France », a tout d'un coup d'éclat. Mais acheter des sièges dans l'espace ne fait pas une politique spatiale. Derrière le...
Paleontologists have identified a new genus and species of pan-shinisaur lizard from a partial upper jaw discovered in southern France, pushing the presence of its lineage in Europe back by at least 30 million years.