Although the formal study of machine intelligence traces its academic roots back to the 1950s, the field experienced a seismic shift in 2022. The release of OpenAI’s ChatGPT—the first large language model to achieve widespread public accessibility through an intuitive, free web interface—triggered a rapid transformation across global industries. The antiques market, an ecosystem traditionally characterized by its reverence for history and a cautious approach to technological disruption, initially stood on the periphery. However, the influence of artificial intelligence has since permeated auction houses, museums, and private collecting circles. As is often the case in legacy sectors, those who adapt to these tools are finding new ways to streamline their operations, while those who remain skeptical are forced to contend with an increasingly automated landscape. What do industry specialists truly think about the growing role of AI in their highly tactile, expertise-driven work? How are major auction houses and museums integrating these systems into their workflows, and where do they identify the most pressing ethical concerns? For the individual collector, where can AI genuinely assist in the pursuit of rare objects, and where does human expertise remain an irreplaceable cornerstone of value? Read Also: Kartell: The Italian Design Icon That Turned Plastic Into High Art Beyond the Icons: Rediscovering the Visionaries of Interwar Art Deco The impact of AI in the antiques sector is defined less by the technical specifications of the software itself and more by the distinct professional contexts in which it is applied. Dealers, auction specialists, and museum curators encounter AI through varying pressures, institutional expectations, and ethical frameworks. This creates a fragmented, uneven landscape—one that is at times transformative and at others deeply contentious. What has become abundantly clear is that AI is no longer a peripheral novelty; it is being woven into the daily mechanics of the trade, even as its limitations and risks remain a subject of active debate. The Dealer’s Dilemma: Efficiency Versus Authenticity For the independent dealer or the small gallery team, AI has emerged as a practical, time-saving tool for managing the administrative burden of the trade. Tasks that once required days of manual labor—such as researching provenance, drafting catalog descriptions, and managing social media content—can now be condensed into a few hours using tools like Copilot. Dealers are increasingly utilizing these models to generate initial catalog notes by inputting basic object descriptions and images. However, this newfound efficiency comes with a significant caveat: the risk of uncritical reliance. Miscataloging caused by unchecked AI output has become a recognized challenge. In the antiques world, where value is often dictated by subtle distinctions in material, technique, or maker, a machine’s inability to grasp nuance can be costly. AI models have been known to misread glaze types, conflate regional carving styles, or incorrectly attribute workshop production to specific, high-value artists. The primary ethical concern here is not simply the occurrence of error, but the question of accountability. Dealers bear the responsibility of ensuring that AI-generated text does not serve as a substitute for true connoisseurship. There is a lingering fear that as more dealers rely on the same algorithmic tools, catalog language will become homogenized, potentially flattening the distinct, expert "voice" that traditionally distinguishes one dealer’s eye from another. Speed, while commercially attractive, must not be allowed to erode the interpretive care and rigor that underpin trust in the antiques market. Auction Houses: Navigating the Market of the Machine In the world of high-stakes auction houses, the role of AI extends beyond simple workflow optimization and into the very nature of the objects being sold. The year 2025 marked a watershed moment with Christie’s New York "Augmented Intelligence" sale, the first major auction dedicated to AI-generated artworks. The event served as a litmus test for the market’s appetite for technological novelty, but it also exposed deep-seated anxieties regarding legal and ethical ambiguity. The sale faced significant backlash, with more than 6,000 artists signing an open letter arguing that the featured works relied on models trained on copyrighted material without proper attribution or permission. Despite the controversy, the sale achieved a total of $728,784. Interestingly, the lack of subsequent high-profile auctions of this nature suggests that the commercial sector remains hesitant about how to position AI-generated objects within a market historically built on pillars of authorship, provenance, and human artistic intent. That is not to say the market has closed its doors to the digital. Individual AI-generated works have achieved notable success; in 2024, "A.I. God. Portrait of Alan Turing," a mixed-media canvas created by the humanoid robot Ai-Da, sold at Sotheby’s for $1.08 million. This remains the highest price ever achieved at auction for an AI-generated portrait, highlighting the market’s unpredictable response to machine-made creative output. Behind the scenes, auction houses are quietly deploying AI to assist with the mundane, yet essential, tasks of valuation and market analysis. AI systems are currently used to analyze past sales, assist with comparable pricing, and support marketing teams in audience segmentation. Even so, specialists remain acutely aware that true valuation and deep connoisseurship rely on knowledge built through the physical handling of thousands of objects—a tactile experience that AI, currently, cannot replicate. Museums and the Preservation of Digital History Museums, operating under different mandates than commercial entities, view AI through the lenses of preservation, access, and institutional responsibility. Their priorities are centered on the interpretation of history and the expansion of public access. In institutions such as the Museum of Tomorrow in Rio de Janeiro, AI has been deployed to improve accessibility, providing real-time sign-language translation and personalized audio descriptions for visitors. Meanwhile, at the Victoria and Albert Museum (V&A) in London, machine learning is being used to conduct critical reviews of historical cataloging. By identifying and flagging problematic terminology in long-standing records, the museum is participating in a broader, necessary ethical conversation about language, representation, and the decolonization of institutional archives. Furthermore, the intersection of AI and collecting is forcing institutions to look toward the future. Digital artifacts are the antiques of tomorrow. In 2023, the V&A acquired a July 2022 issue of The Economist featuring a cover generated by an early version of the Midjourney program. This acquisition captures a distinct cultural moment where intangible digital processes were first becoming embodied in tangible objects. It poses a difficult question for future curators: how does an institution preserve the context of an AI-generated object, and how will future researchers interpret items created by systems whose training data and internal logic may never be fully transparent? Industry Expert Insights Otto Billström, Head of Modern Decorative Art & Design at Bonhams London, provides a pragmatic view of how these tools function within a top-tier auction house. For Billström, large language models like ChatGPT, Copilot, and Gemini are useful as a "first step" for identifying well-known artists or historical marks. However, he remains adamant about their limitations. "While it can recognize basic shapes and patterns, the tools are far from refined enough to tell makers who work in similar styles apart and will usually err on the side of a more well-known maker," Billström explains. "Similarly, it is not capable of telling a genuine work apart from a fake. It is therefore important to always do your own research and arrive at your own conviction." Billström emphasizes that AI cannot replace the "bedrock of the market"—the human relationships, historical context, and lived expertise that define the antiques ecosystem. When it comes to valuation, he points out that AI often lacks the ability to distinguish the context of a sale. A prime example is the 2026 Christie’s sale of "MASTERPIECES: The Private Collection of S.I. Newhouse." The collection of the late publishing titan realized over $630 million, a result driven as much by provenance and historical prestige as by the individual objects themselves. "An AI will simply look at the price achieved," Billström notes, "meaning in normal market conditions, the result may be anomalous and therefore create false expectations." Utilizing AI in the Private Collecting Sphere For the private collector, AI serves as a powerful, albeit cautious, assistant. It can be a highly effective tool for forming first impressions and organizing a personal collection. Image-recognition software can offer initial suggestions regarding an object’s maker, period, or style based on a photograph. While this is helpful at a flea market or when browsing online listings, these findings must be treated as suggestions rather than definitive facts. Beyond identification, AI can assist in deepening a collector’s historical understanding. It can summarize complex stylistic features, explain the differences between related regional crafts, or translate historical marks and inscriptions. It acts as an interpretive guide, helping beginners navigate unfamiliar categories of collecting. Furthermore, it is a practical tool for archival purposes; AI can help transform scattered photographs and notes into consistent, well-organized catalog entries. However, as in the professional trade, accuracy remains paramount. All AI-generated information must be verified against scholarly sources before it is committed to a permanent record. The antiques world is currently navigating a complex intersection between tradition and digital modernity. While AI accelerates research and expands access, it also exposes the enduring necessity of human judgment, material literacy, and cultural interpretation. As institutions and individuals continue to negotiate the limits of this technology, AI is becoming less of a disruptive force and more of a lens through which the trade is forced to reconsider the nature of expertise. Ultimately, the presence of AI serves as a reminder that innovation in the antiques sector is most meaningful when it strengthens, rather than shortcuts, the rigorous knowledge that has defined the field for generations. Wiktoria Kijowska, a design historian and graduate of the V&A/Royal College of Art History of Design program, continues to bridge the gap between historical narrative and contemporary digital practice through her work in curatorial research and digital storytelling. Her ongoing engagement with the intersection of antiques and modern technology reflects a growing trend in the sector: the desire to keep historical narratives vibrant and accessible in an increasingly automated age. Post navigation The Evolution of the Ice Bucket: From Rare Luxury to Tabletop Icon