Hephaestus is the world’s only company that can deploy artificial intelligence to identify artworks with greater than 98.2% accuracy, as was achieved in a recent case study comparing the work of Canaletto and Bellotto. Using high-resolution images captured from a medium-format camera, Pictology is trained on a homogenous data set with fixed variables, eliminating the need for onerously large data sets that would be difficult to both acquire and curate.
The results are so conclusive that we are working with an AA-rated insurance company to issue the first authentication insurance product on paintings. While the implications of this development are profound, in this article we will focus briefly on Hephaestus’ use of artificial intelligence algorithms to solve classic problems in art history and attribution -- one significant way in which Hephaestus distinguishes itself from its competitors.
Art forgers are often underestimated and sophisticated forgers can be amongst the most knowledgeable people on the planet with respect to technology and technical art history. Scientific tests can be reverse-engineered: historic pigments can be recreated or reused, and carbon dating measurements can be contaminated and manipulated. Yet, these are the methods most often employed in the art market. This underestimation stems from a more general misconception that forgers operate solely through manual replication, disregarding their oftentimes deep understanding of both historical techniques and modern advancements. In reality, many forgers possess an intricate knowledge of art history, materials, and techniques, allowing them to produce convincing replicas that can deceive even experienced connoisseurs.
It could be argued that there is a conflict of interest insofar as the fact that the same people selling artworks are also the authenticators.
The situation is analogous to what occurred in the Global Financial Crisis of 2009. It was not so much a liquidity crisis as it was a crisis of trust. The “referees” in this case were the rating agencies, Standard & Poor’s and Moody’s, who were paid by the investment banks to rate collateralised loans. When the referee said “yes” everyone at the table was off the hook: investors got to buy AAA-rated securities, investment bankers made their fees, rating agencies were paid, and unqualified buyers got to borrow money.
The 2023 Deloitte article crystallised authenticity as a major concern not only for collectors but also for wealth managers, as discussed in our recent insights article.
Hephaestus was founded with the mission of eliminating forgery in the art market, which is only possible with the removal of shortcomings of scientific tests and conflicts of interest. By applying stringent forms of testing for the material aspects of paintings and applying them in a scientific sequence such that each test increasingly eliminated, a far more accurate and reliable set of scientific protocols was developed. Nonetheless, the scientific analysis of materials can only ever prove a painting is a forgery. The determination of authorship and attribution has always been a thorny subject.
Authentication in the art world is often likened to a three-legged stool, supported by provenance, connoisseurship, and scientific analysis. However, the advent of artificial intelligence (AI) introduces a fourth leg to this metaphor. Provenance provides crucial documentation of an artwork's ownership history, connoisseurship offers expert insight into an artist's style and technique, scientific analysis delves into the physical composition of the artwork and AI adds a new dimension.
Artificial intelligence brings computational power and pattern recognition capabilities to the authentication process, enhancing its accuracy and efficiency. By analysing vast amounts of data and identifying subtle patterns, AI algorithms can assist in determining an artwork's authenticity with unprecedented precision. Hephaestus' AI art detection algorithms can learn from data sets of authenticated artworks to recognise unique characteristics and attributes associated with specific artists or styles. This enables our AI to detect anomalies, inconsistencies or potential forgeries that might elude human experts or traditional scientific methods. As a result, the inclusion of AI in our protocol expands the scope and reliability of the authentication process, providing a more comprehensive and robust framework for verifying the authenticity of artworks. While human experts may rely on subjective judgments and personal biases, AI algorithms can provide consistent and evidence-based assessments, reducing the risk of human error and bias in the authentication process.
Enter machine learning and Hephaestus’ crossover with astrophysics. The mathematical inferences of large data sets of light have given scientists a greater understanding about the origins of the Milky Way. If astronomers could use artificial intelligence to develop “light signatures” from data sets about objects that were light years away, why couldn’t similar forms of analysis be used to develop “signatures” for the brushstrokes evident in paintings?
Working with preeminent astronomers at Princeton University, the University of London and the Australian National University, Hephaestus used “scattering transform” to detect more than one million dimensions and reduce them to about a 100-dimensional “fingerprint”. Some of these dimensions include the pressure with which a paintbrush is applied to a canvas, the average distance between objects, the spacing of strokes and the presence or absence of shapes in the artist’s oeuvre. If you will, Hephaestus' Featuriser extracts these characteristics for each artist, assigning each one a value that assesses its contribution to the artist’s style. A three-dimensional contour is generated and the degree to which a painting fits inside the contour determines the likelihood that it may be attributed to a particular artist. Art forgers, who are becoming increasingly sophisticated in their techniques, challenge traditional art authentication methods, and Hephaestus recognises and actively addresses this.
It is not only our competition that cannot imitate our art authentication tests; neither can forgers. Two critical outcomes of Hephaestus’ tech are: (1) these dimensions are not visible to the eye and may only be detected by statistical analysis of millions of data points concurrently (2) they are therefore nearly impossible to forge. We have not yet dealt with the case of a living artist “forging” their own works.
An illustration may demonstrate the effectiveness of the technique. A classic problem in art history and connoisseurship is distinguishing the works of Canaletto from his nephew and apprentice, Bellotto. Hephaestus’ algorithms solved the problem with 98.2% accuracy.
Brian Allen, former director of the Paul Mellon Centre for Studies in British has discussed how the difficulty for connoisseurs, and especially young art historians, to distinguish between the work of Canaletto and Bellotto is symptomatic of the problems associated with the shift in art history education towards social history which began in the early 1980s. One of the fundamental consequences of art historians being unable to distinguish between the hand of a master, an assistant or a copyist is that forgeries will be missed, says Allen. This is one of the key reasons why artificial intelligence is becoming an increasingly useful tool in Hephaestus' art authentication protocol; it is able to identify the fine nuances of specific artists and Masters.
The “devil in the details” is the collection of accurate data as the use of forged works or misattributed ones will dramatically lower the accuracy of what may be detected. Hephaestus' database of artists is growing slowly and painstakingly, but the results are astonishing. These are the highest evidentiary standards in the world of art authentication; our certificate of authenticity protects buyers, sellers and lenders to enable a more transparent art market.
The integration of advanced technologies continues to transform not just the realm of art authentication, enhancing traditional methods with unparalleled precision and effectiveness, but the art market as a whole. Digital imaging technology empowers experts to scrutinize artworks at a microscopic level, uncovering hidden layers or alterations indicative of forgery while, Pictology, Hephaestus' AI art authentication algorithms, trained on extensive databases of authenticated artworks, possesses the potential to discern patterns and identify potential forgeries with remarkable accuracy.
For sellers, a conclusive authentication ensures the ability for a given to artwork to achieve the highest possible sale price and, for buyers, authentication serves not least as a guarantee of value. Authentication also plays a crucial role in scenarios such as tax-deductible donations, estate valuations, and legal proceedings like divorces and partnership dissolutions. In today's art world, authenticity is no longer solely reliant on subjective opinions but, rather, on extensive research, references, analyses, provenance, and forensic examinations.
AI plays a key role in Hephaestus' art authentication protocol, given its capacity to identify the hand of an artist and their signature qualities. Therefore, used in tandem with scientific analysis, provenance research and connoisseurship, AI provides the conclusive element of Hephaestus' authentications which, from early 2024, will be protected by an authenticity insurance product.
Hephaestus: eliminating forgery from the art markets, one painting at a time.