AI in art market: implications for appraisers

The fine art market has traditionally been slow to adopt new technology. However, with the emergence of AI, approaches to authentication, valuation and fine art investment are showing signs of disruption.

Insurance and financial services firms are investing heavily in customised AI tools to assist their employees by automating recurring tasks. In the art world, appraisers are beginning to see that these efforts include the use of predictive fine art and collectible valuation applications for policy coverage and estate planning.

The effective use of AI-driven price predictions and market performance analytics requires training, curated datasets and the ability to design precise prompts that yield meaningful results. Interpreting results from AI-powered information requires fine art expertise and advanced knowledge of a property's value characteristics in a dynamic marketplace.

Valuation remains an area where the expertise of appraisers plays a vital role in protecting the long-term needs of fine art collections and the interests of collectors, museums and artists.

Used strategically, AI can improve productivity in artist research, provenance investigation, comparative analysis and descriptive cataloguing.

For example, in the luxury goods sector, where objects share technical specifications, AI can provide reliable descriptions and comparisons.

For the fine art market, image-recognition functions may help identify signatures, monograms or stylistic parallels. AI is also proving useful for the translation of foreign-language catalogue entries and publications, and inscriptions appearing on the recto or verso of works of art.

Beyond research, AI can assist with writing tasks such as editing and improving the readability of reports or marketing materials.

Appraisers can use AI to streamline administrative processes in service quotes and proposals. AI can also process large volumes of sales data to reveal patterns, helping to analyse current market trends.

At present, the overall fine art market is in decline, while the broader luxury goods market – vintage cars, handbags and designer jewellery – is expanding. These objects, produced in multiples with standardised features, are easier for AI to assess. 

Works of art, by contrast, are inherently unique. AI can catalogue and compile factual attributes such as design specifications, digitised reference materials, artists' biographies and published sales results.

However, AI cannot interpret the interplay of authorship, medium and artistic movements. Nor can it replicate the role of a seasoned dealer who generates demand through taste, reputation and influence.

This caveat illustrates that while AI may offer efficiencies, it is accompanied by significant drawbacks.

With increased use in a changing market, appraisers must understand the capabilities of AI, but also the associated risks.

Privacy is another risk. Inputting proprietary information – such as provenance details, ownership records or images – into AI tools could potentially expose sensitive client data and affect a client's trust in their appraiser.

Authentication also presents particular challenges for AI. Although the technology can analyse patterns like brushstrokes, image-based analysis is limited when datasets consist of inconsistent or degraded scans of artworks.

For reliability, images would need to be high-resolution, first-generation scans produced under museum-standardised conditions – an impractical requirement at scale.

AI tools are also unable to distinguish between base materials such as canvas, paper or wood panel, which are important market distinctions.

Instead, fine and decorative arts appraisers must still rely on various sources of information gathered by industry relationships to capture unpublished sales and market intelligence, while avoiding reliance on AI-generated information as the definitive source.

Datasets themselves pose additional problems. Metadata is often incomplete or biased, particularly with regard to underrepresented artists.

This algorithmic bias can perpetuate existing market inequalities by privileging artists already dominant in the historical record or concentrated in major art centres such as London or New York.

Bias can also appear in the prioritisation of certain media based on perceived value – for instance, highlighting paintings over sculptures or digital works.

For collectors using AI to select artworks, this can result in emerging or rediscovered artists being overlooked, simply because their online presence is limited.

Similarly, online sales data may reflect the artist's recognition in one country, producing price suggestions that may not be achievable elsewhere.

Proprietary valuation programs illustrate further limitations with AI. Auction results, while public, represent only part of the market. Secondary market sales at auction may not be the best market for the purpose of the appraisal and the desired type of value.

In the primary market, private sales, undisclosed final prices and gallery transactions remain invisible to most datasets. Even when asking prices appear online, the true sale price is often unavailable.

Independent appraisers also face a structural disadvantage as insurers and financial institutions invest in bespoke AI tools.

Unlike corporate institutions that can fund custom AI development, most appraisers work as consultants or in small practices, and therefore won't have access to the most cutting-edge AI technology and datasets.

In addition, AI is increasingly used to predict an artist's market trajectory or identify potential investments.

However, such predictions often fail to account for variations in medium, shifts in artistic style or the complex social and cultural factors that drive collector behaviour.

In order to stay competitive, it is also important that appraisers demonstrate fluency in human-machine collaboration, while recognising bias and ensuring their clients receive opinions of value suited to the appraisal's intended purpose.

AI offers important opportunities, but it is not an authority in the fine art market. While some believe AI will bring transparency and accessibility to the art market, its conclusions depend on flawed and incomplete datasets.

Recognising issues with AI while continuously watching for market fluctuations will ensure that fine art appraisers continue to be active members of professional service teams, advising clients in good times and bad.

For appraisal clients, AI-generated information can serve as a useful guideline to help narrow down acquisition options, highlight works requiring deeper investigation or support preliminary pricing discussions.

However, appraisers should emphasise to their clients that AI-only results provide starting points rather than final answers.

Best practices that appraisers should adopt involve pre-selecting datasets from reputable sources and carefully engineering AI prompts.

Citations and the disclosure of AI use, along with a clear statement of its limitations, should become standard practice among art appraisers to maintain credibility.

Without such controls, AI can pull from inaccurate, biased or misleading sources, triggering poor business decisions and increasing appraisers' liability.

Any information that AI tools produce must be verified by an appraiser before being included in professional reports.

Appraisers in all specialties should check with their professional associations for guidance and new standards around the responsible use of AI.

For professional service providers – such as estate and financial planners – or for collectors seeking just-in-time authentication, valuations or investment predictions, reliance on these tools alone remains risky.

The art market continues to depend on experience, social networks and professional judgement – qualities that AI cannot replicate.

Despite growing interest in AI, connoisseurship remains central to fine art appraisal. At present, AI lacks the capacity for critical thinking, empathy and ethical behaviour – qualities that underpin professional appraisal practice.

Evidence-based appraisal reports adhering to recognised appraisal standards remain the foundation for informed decision-making, and qualified appraisers are still indispensable to the business of fine art.