Why it matters
Without it, misdescribed lots, inconsistent estimates, typographical errors, and style deviations undermine trust, complicate buyer expectations, and slow post-sale settlements.
From design to safety
Design
AI-enforced style and consistency planning
AI analyzes existing catalogues, house style guides, and term policies to outline mandatory data fields, naming conventions, and proofing rules for future catalogues and terms, including pagination standards and image attribution rules.
- catalogue copy desk
- lot number register
- estimate book
- condition report archive
Build
Automated content ingestion and proofreading
During intake and photography, AI checks catalogue text against source data, flags typos, ensures alignment with house style, and records corrections, including cross-referencing estates and provenance notes for accuracy.
- catalogue copy desk
- lot number register
- estimate book
- condition report archive
Operation
Live proofreading during publication
AI scans evolving proofs in real time, detects inconsistencies across entries, and triggers prompts for human review before final publication, while logging accepted changes for audits and future training.
- catalogue copy desk
- lot number register
- estimate book
- publication proofs cabinet
Maintenance
Ongoing quality governance
AI tracks variance against historical data, updates style rules, reconciles changes across batches, and validates new catalogues against master templates before final roll-out to sale platforms.
- catalogue copy desk
- lot number register
- estimate book
- condition report archive
Safety
Compliance and provenance checks
AI flags potential AML or provenance concerns in catalogued items, ensuring adherence to compliance protocols, export controls, and embargo rules, with escalation to the Compliance Officer for sign-off.
- catalogue copy desk
- lot number register
- estimate book
- house style guide
Part by part
- 1
catalogue copy
- Watch for
- Inconsistent terminology, missing field values, or misapplied house style across related entries, including title formatting, date spellings, and provenance phrasing.
- AI action
- Flag inconsistencies, propose corrected wording, auto-suggest standardized phrasing, and attach cross-references to source documents.
- 2
lot number register
- Watch for
- Duplicate lot numbers, mislinked serials, or missing cross-references to estates, with potential estate-ownership gaps.
- AI action
- Detect duplicates, warn on mismatches, generate cross-reference map, and verify estate IDs against consignor records.
- 3
estimate book
- Watch for
- Estimates out of band with similar lots, missing currency, or incorrect reserve status, including currency mismatch and typographic errors in decimals.
- AI action
- Normalize ranges, verify currency, flag reserve inconsistencies, and attach market notes for context.
- 4
condition report archive
- Watch for
- Missing condition notes, inconsistent grading descriptors, or nonstandard terminology, including damage terminology and colour descriptors.
- AI action
- Suggest standardized condition terms, fill gaps using image analysis, and attach reference image snapshots.
- 5
house style guide
- Watch for
- Unaligned typography, capitalization, or terms across multiple catalogues, including bulleting and header hierarchies.
- AI action
- Enforce style rules, propagate corrections to all active records, and timestamp revisions for audit trails.
- 6
publication proofs cabinet
- Watch for
- Unpublished proofs with conflicting revisions or missing approval signatures, including delayed sign-offs.
- AI action
- Identify pending approvals, route to reviewer queue, and generate compliance summary for final release.
Worked example
Catching a seal stamp in the stacks
Illustrative scenario with sample data.
A regional auction house wants to reduce cataloguing inconsistencies across five sale weeks and ensure uniform style before print.
Signal in
annualLotCount: 25000; categories: [Paintings, Sculpture, Decorative arts, Prints, Watches]; dataFields: [title, artist, date, dimensions, materials, provenance, estimate, reserve, lotNo, condition]
System out
Duplicates removed, style rules applied, and corrected entries published in final proofs; flagged items with provenance concerns.
Human sign-off: Head of Cataloguing approves the final proof batch and signs off on compliance checks.
How to roll it out
Phase 1
Weeks 1 to 2
Baseline ruleset established; existing catalogues indexed for style deviations and data gaps.
Phase 2
Weeks 3 to 6
Automated checks integrated into intake and production; alerts routed to cataloguers.
Phase 3
Weeks 7 to 12
Full catalogue publication workflow with AI-assisted proofs and human sign-off.
Where people stay in charge
- All AML, provenance, and export-related flags must be reviewed and signed off by the Compliance Officer before any public release.
- Critical data changes require two-person verification from a senior cataloguer and a senior editor.
- Final approval of house style deviations must be obtained from the Head of Cataloguing or Managing Director.
Questions owners and crews ask
How does AI proofreading reduce errors in lot descriptions and estimates?
AI checks for consistency across fields, flags omissions, and standardizes phrasing; human editors review flagged items to ensure accuracy and compliance before publication.
Who approves corrections identified by the system?
Corrections are approved by senior cataloguers and the head of cataloguing, with final sign-off from the managing director for policy alignment and record-keeping.
Can this system handle legacy catalogues with inconsistent terminology?
Yes, it maps legacy terms to the house style, flags residual inconsistencies, and guides editors to unify terminology across all records, including archival notes.
What data privacy and provenance controls exist?
Provenance data is checked against compliance rules; AML flags are escalated to the Compliance Officer for regulatory review and export eligibility decisions, with audit trails.
Sources and further reading
- RICS: Profession StandardsRICS professional standards on valuation and estimate accuracy for regulated auctioneers.
- Society of Fine Art Auctioneers and ValuersFine art auctioneers' code of practice covering catalogue descriptions and house conventions.
- GOV.UK: Money Laundering Supervision for Art Market ParticipantsHMRC anti-money laundering registration, customer due diligence and record keeping rules for art sales.
- GOV.UK: Exporting or Importing Objects of Cultural InterestUK export licence rules for cultural goods, value thresholds and import requirements.
- CINOADealer due diligence and trade regulation guidance for art and antiques businesses across member associations.
- UNESCO: Fight Against Illicit Trafficking of Cultural PropertyUNESCO guidance on illicit trafficking, art market due diligence and restitution of objects.
- Getty Research Institute: Provenance ToolsArchival sales catalogues and historic ownership records for checking provenance notes against sources.
- The Art Newspaper: AuctionsAuction sale reports covering results, guarantees and auction house strategy.
- ARTnews: MarketAuction results, sale previews and market data to cross-check estimates against comparable sales.