The Next Great Forger: Art Market Faces New Risks As AI Mimics Authentic Records

As artificial intelligence quietly enters the art market, its influence is extending beyond creating images to forging documentation—invoices, certificates of authenticity, and ownership histories. This new form of digital forgery is challenging insurers, brokers, and art scholars, raising fundamental questions about trust and verification in a sector long reliant on provenance.

When Documentation Becomes Vulnerable

For centuries, provenance—detailed records tracing a work’s ownership—has been a key defense against art fraud. Catalogs, certificates, and invoices have carried as much authority as the artworks themselves. However, recent reports indicate that AI-generated documentation is increasingly appearing in sales and insurance claims.

According to specialists, these AI-crafted files are often convincing at first glance: polished language, proper formatting, accurate-sounding dates, and seemingly credible gallery references. Yet subtle inconsistencies—metadata irregularities, stylistic errors, or fictitious institutions—can reveal the underlying falsity. Detecting such manipulations is becoming increasingly difficult as generative models grow more sophisticated.

The Persuasive Power of AI

Experts describe AI’s role in documentation not as deliberate deceit, but as a form of systemic plausibility. Large language models are designed to provide confident responses even when they lack factual grounding, a phenomenon known as “hallucination.”

In practice, this has led to instances where collectors and insurers rely on AI to verify ownership or valuations, only to receive fabricated yet seemingly credible results. Once introduced into a claim or transaction, these falsified records can be extremely challenging to disentangle from legitimate documentation.

Implications for Insurers and Brokers

The insurance industry, which depends heavily on accurate documentation for risk assessment and claims processing, is feeling the strain. Grace Best-Devereux, a claims adjuster at Sedgwick, highlighted how highly plausible AI-generated documents are complicating fraud detection. Similarly, Olivia Eccleston, a fine art broker at Marsh McLennan, noted that chatbots are being used to forge sales invoices, valuations, and certificates of authenticity. In one instance, a review of a loss claim revealed dozens of documents for a painting collection that initially appeared credible, but metadata analysis later suggested the entire collection may have been fictitious.

A Modern Twist on an Old Practice

Forgery is not new to the art world. Historical examples include Michelangelo’s 1496 sculpture alteration, where the young artist used chemical treatments to make a piece appear ancient. Today’s difference lies in speed, scale, and automation: generative AI can produce countless variations instantly, with minimal effort and virtually no human oversight.

The challenge for the contemporary art market is no longer just spotting fake artworks—it is navigating a digital documentation ecosystem where trust itself can be artificially generated.