Artmarket.com is delaying the public rollout of new artificial intelligence features at its Artprice unit as it replaces parts of its core technology with an “AI-first” architecture, a more extensive rebuild than the company’s earlier plan to add AI components gradually.
The French art-market data group said its board approved the strategic shift after reviewing the scale of work required to deploy proprietary systems called “Intuitive Art Market®” and “Blind Spot®.” Artmarket said the revised implementation schedule does not alter its stated financial trajectory, though it did not provide a new date for a customer launch.
The move places internal systems, data quality and testing ahead of a rapid product release. Artprice plans to first equip staff with dedicated AI hardware and edge-computing units, then rewrite its processes for collecting, standardizing and enriching auction data to work with deep-learning models and proprietary algorithms.
That approach reflects a practical constraint in the art market: predictive software is only as reliable as its underlying sales records, artist attribution, provenance information and cataloguing standards. Artprice’s business depends on turning often fragmented auction-house information into comparable records across countries, currencies, languages and historical periods.
Artprice rebuilds its data platform before customer release
Artmarket said the first stage of the overhaul will focus on internal deployment and recalibration of staff workflows. The company plans to subject its AI tools to beta testing and stress tests before offering the resulting services to clients.
Only after that process will Artprice release a database platform built natively as an AI-first system, the company said. Rather than placing generative tools on top of a conventional search database, the planned platform would combine standardized and certified data, deep learning and algorithm security from its initial design.
The company said its systems will be able to operate in more than 40 languages while retaining artwork titles in their original form. That detail matters in a market where translations can obscure the distinction between an artwork’s official title, an auction-house description and a catalogue raisonné reference.
Artprice also plans to make its interface capable of proposing follow-up questions based on a customer’s initial natural-language query. Such a function could help users narrow broad requests — such as an artist’s market performance — into questions concerning periods, media, geographic sales patterns or comparable works.
Artmarket said its art-market databases include more than 30 million indices and auction results covering over 915,300 artists. Its Artprice Images® bank contains 181 million digital images of artworks and reproductions dating from 1700 to the present, according to the company. Artmarket also said its services have 9.3 million registered members.
Auction-house network underpins the data strategy
The company said it controls a network of 7,200 partner auction houses for contracted data collection. That network is central to the plan because auction results, estimates, withdrawn lots and sales histories form the raw material for Artprice’s valuation products.
Artmarket cited Gartner Group and Europol Innovation Lab estimates that 70% of synthetic data was “uncontrolled” as of June 30, 2026, presenting the figure as evidence of information-quality risks across the open internet. The company’s response is to emphasize contractual data sources and controlled processing rather than relying on freely scraped online material.
Its planned commercial model includes predictive APIs and embedded inference systems for institutional workflows. APIs, or application programming interfaces, allow one company’s data or software outputs to be used within another organization’s own platform. In practice, Artprice is positioning its AI tools for clients that want valuation signals or market data delivered directly into internal research, cataloguing or advisory systems.
Artprice listed annual subscription prices of $1,600 to $2,500, also stated as €1,600 to €2,500. The company did not specify whether the AI platform and APIs will carry separate pricing or how access will be structured across customer groups.
Blind Spot focuses on gaps between auction sales
Artmarket described “Blind Spot©” as a system originally designed to model price discontinuities between two public auction sales. The tool uses an artist’s global index history and the traceability of auction records to assess changes between transactions.
The company used Jackson Pollock as an illustration, citing a work that sold for $4 million in 1998 and $58 million in 2026. A comparison of that kind can be informative, but its value depends heavily on factors beyond an artist-level price index, including the work’s size, condition, medium, exhibition history, provenance, guarantees and the specific auction environment.
Artmarket’s proposed system could therefore be most useful when it makes the underlying comparables and assumptions visible rather than presenting a single automated valuation as definitive. Art-market pricing remains less standardized than liquid securities markets, and two works by the same artist can vary sharply in value even when sold within a short period.
The company also said it has deployed two forms of agent-based AI for data collection and U.S. commercial targeting. Deterministic agents are intended to gather structured information without interpretation through official and contractual partnerships, according to Artmarket. Probabilistic agents are being developed for commercial activity in the United States, where the company said they would adapt to linguistic and cultural conditions across states.
Ehrmann family and Groupe Serveur plan further purchases
Alongside the technology announcement, Artmarket.com said founder Thierry Ehrmann’s family and majority shareholder Groupe Serveur intend to increase their holdings through additional share purchases.
The company said disclosures to France’s Autorité des marchés financiers, or AMF, would be made within applicable legal deadlines and during permitted trading windows. It linked the planned purchases to the substantial spending required for Artprice’s vertical-AI development.
The intended increase in ownership gives the company’s controlling shareholders a more direct financial commitment to a project that requires infrastructure spending before the new products generate revenue. It also signals that Artmarket expects its existing art-market records, image archive and auction-house relationships to remain valuable assets as AI tools make data provenance and reliability more commercially consequential.
Artmarket has framed the overhaul as a modernization of its established data business rather than a separate venture into tokenized art or digital-asset trading. The immediate test will be whether its internal systems can turn a large historical archive into faster, auditable and commercially useful outputs without weakening the data controls on which its market reputation depends.
For deeper insight into AI-driven trading tools, explore our guide to AI copy trading and how it works today.
Disclaimer: The content on this page is provided for general informational purposes only and does not represent the views or financial advice of Toobit. We make no guarantees regarding the accuracy or completeness of this information and shall not be held liable for any errors, omissions, or outcomes resulting from its use. Investing in digital assets involves risk; users should independently evaluate their financial situation and the risks involved. For further details, please consult our Terms of Service and Risk Disclosure.
