Why it matters
Without guided AI-supported selection, irreversible treatment damage, uncertain reversibility, and disputes over conserve-lost details can erode loan eligibility and insurer confidence.
From design to safety
Design
AI-guided material choice concept
The AI reads substrate behavior, historical usage, pigment compatibility, aging trajectories, and reversibility goals to propose gesso fills, plaster cores, epoxy interfaces, and calibrated toning ranges with documented rationale.
- gesso fills station
- plaster base station
- toning workstation
- epoxy resin prep bench
Build
Materials data integration in setup
Data pipelines ingest substrate tests, archival material dictionaries, pigment match records, and conforming solvent interactions to calibrate the AI's compatibility matrix and produce formal pre-treatment sheets for approval.
- epoxy resin station
- cellulose filler station
- carving bench
- toning pigments cabinet
Operation
Real-time matching during treatment
During treatment, the AI proposes fill types, thickness guidance, surface texture steps, and pore-filling depth while logging material choices, handling parameters, cure progress, and substrate responses for future replication.
- gesso fills station
- epoxy resin station
- plaster base station
- toning workstation
Maintenance
Ongoing material performance monitoring
The system tracks aging indicators, solvent interactions, reversibility checks, and micro-abrasion wear to adjust future interventions, update condition reports, and maintain a traceable material history.
- cellulose filler station
- toning pigments cabinet
- carving bench
- gesso fills station
Safety
Ethics and safety governance
A designated conservator validates AI recommendations, confirms reversibility, ensures COSHH compliance, and maintains PACR-accredited oversight with ongoing ethics reviews and documented decision trails.
- epoxy resin station
- plaster base station
- toning workstation
- gesso fills station
Part by part
- 1
gesso-fills station
- Watch for
- risks of translucency mismatch, substrate swelling under humidity, and crepe-layer formation during curing
- AI action
- AI flags mismatch risk, recommends alternative gesso recipe with calibrated water-to-pize ratio and extended curing window
- 2
epoxy-resin station
- Watch for
- exotherm control, shrinkage gaps, and surface bloom in older substrates
- AI action
- AI triggers test batch, records max processing temperature, and adjusts catalyst-to-resin ratio before full-scale application
- 3
plaster-base station
- Watch for
- crack bridging impedance, cure shrinkage, and core mismatch with substrate
- AI action
- AI recommends porosity adjustments, lime-putty mixes, and pore-filling depth based on substrate hardness and historic repair style
- 4
cellulose-filler station
- Watch for
- fiber pull-out, adhesive incompatibility, and swelling mismatch
- AI action
- AI analyzes compatibility matrix with resin types and suggests alternative cellulose grade and pre-treatment steps
- 5
toning workstation
- Watch for
- color-matching drift, glaze cohesion, and surface gloss variance
- AI action
- AI computes pigment toning ratios, patch blend plan, and cross-checks with aged varnish layers for tonal harmony
- 6
carving bench
- Watch for
- tooling marks removal vs. surface texture loss, depth control
- AI action
- AI guides texture-preserving shaping parameters, records brush and scalpel pressure profiles, and logs tactile feedback for future calibrations
Worked example
AI-guided gap-fill selection for a 19th-century wood-leaning statue
Illustrative scenario with sample data.
A private studio with a four-person team evaluates cracks and losses on a carved figural sculpture; AI assists in selecting fill materials that balance reversibility, tonal fidelity, and archival traceability.
Signal in
substrate: larch wood; existing fills: shellac-based; cracks: 4–6 mm; pigment data: era-appropriate ochres; humidity: 45%; tests: solvent compatibility; records: condition reports 2024-03; previous repairs: hide glue consolidation 2012
System out
Recommended fills: gesso base with proprietary plaster core; toning pigments list; curing times; treatment proposal with reversibility notes; updated condition report template; material sheet detailing adhesion tests and removal plan.
Human sign-off: Head conservator approves the treatment proposal, signs the PACR-compliant reversibility plan, and authorizes the use of the proposed fill system in line with ethics expectations.
How to roll it out
Phase 1
Weeks 1 to 4: Baseline
Establish data pipelines, calibrate material compatibility matrix, and create initial treatment proposal templates for early cases.
Phase 2
Weeks 5 to 12: Pilot
Run parallel fills on test panels, refine AI thresholds, document material performance, and generate formal treatment reports for clients.
Phase 3
Weeks 13 onward: Scale
Full integration into casework, ongoing calibration, and routine updates to condition archives with versioned records.
Where people stay in charge
- Only a qualified conservator or PACR-accredited supervisor signs off on fill-material recommendations and reversibility notes.
- AI suggestions must be reviewed against reversibility requirements, documented in the treatment proposal, and kept in the condition archive.
- All COSHH and safety data must be validated by the designated safety officer and cross-checked in the risk assessment before application of any fill material.
Questions owners and crews ask
What data does the AI read to select gap-filling materials?
The AI reads substrate type, existing fills, pigment compatibility, aging behavior, humidity, solvent interactions, historical treatment records, and reversibility history to propose materials and sequence with auditable justification.
Who approves AI-generated treatment proposals?
A PACR-accredited conservator reviews, signs off on the treatment proposal, ensuring reversibility, ethics, and safety compliance, and the lead conservator maintains final sign-off authority.
How is reversibility ensured in AI-guided fills?
The AI prioritizes reversible materials, records removal pathways, and provides a documented removal plan; the conservator validates feasibility and seals this in the formal treatment proposal.
What records does the system generate?
It creates a treatment proposal, a materials sheet, a procedural log, a condition report update, an after-care plan, and a complete audit trail for future conservators.
Sources and further reading
- Icon: The Institute of ConservationIcon explains PACR accreditation requirements and the conservation register standards.
- HSE: COSHHHSE guidance on control of substances hazardous to health, solvents and exposure assessments.
- ICCROM: Preventive ConservationAgents of deterioration and risk management methodology for preventive conservation planning.
- ICOM-CC Working GroupsSpecialist groups covering paintings, paper, and metals materials research and treatment practice.
- Getty Conservation InstituteConservation science research on materials, imaging, and collection environment interactions.
- Library of Congress: Preservation SciencePreservation research, materials analysis, and testing methods for reversible treatment evaluation.
- American Institute for ConservationUS conservation body's ethics code and specialty group guidance for practitioners.
- Tate: ConservationTate publishes conservation research on sculpture materials and treatment techniques for practitioners.
- HSE: Local Exhaust VentilationFume extraction design and testing requirements for workplaces handling resin and solvent vapours.
- ICOM Code of Ethics for MuseumsICOM ethical standards for care, conservation, and restoration of museum collections.