Искусственный интеллект: новый инструмент мошенников в мире искусства English translation: Artificial Intelligence: A New Tool for Fraudsters in the Art World

Artificial intelligence is aiding fraudsters in forging provenance documents and certificates of authenticity for art pieces, according to insurance broker Olivia Eccleston, as reported by the FT.

She noted that this trend has added a new dimension to the age-old issue of forgery and fraud in the art market.

One loss adjustment expert revealed that he received dozens of valuation certificates as part of an insurance claim for a collection of decorative paintings that initially appeared credible. However, a mistake made by the claimant led the analyst to suspect forgery. It turned out the documents were generated using an automated system, with the description field for each piece being identical.

Other industry representatives highlighted that AI has sometimes been used with malicious intent, as fraudsters attempt to fabricate provenance documentation. There have also been incidents where users requested information about their artworks from neural networks, only to receive hallucinatory responses.

Angelina Giovanni, co-founder of Flynn & Giovanni, noted that AI makes it easy to create counterfeit documents due to its operational mechanics. Since the neural network must produce an answer, if provided with enough information, «somewhere the LLM will make a guess.»

Experts emphasized that the use of artificial intelligence represents a new phase in age-old practices. While previously individuals would steal or forge stationery with the logos of reputable institutions, now they are utilizing cutting-edge technology.

«I have seen many forged documents throughout my career. This is not a new phenomenon, but AI has made it more realistic,» stated Filippo Guerrini-Moraldi, head of the fine art department at Howden Insurance.

Harry Smith, executive chairman of Gurr Johns, mentioned that AI has simplified and accelerated the work of fraudsters.

To combat this emerging issue, it is possible to check metadata in digital documents for signs of AI intervention. This approach is employed by Grace Best-Deverex, a loss adjuster at Sedgwick.

Her colleagues follow a similar strategy, using LLM tools to verify document authenticity. However, this task has become increasingly challenging due to the ongoing development of the technology.

It is worth noting that in the 2024 fiscal year, the U.S. Department of the Treasury used artificial intelligence to analyze data and recover over $4 billion obtained through fraud and improper actions.