When I tell fellow sculptors I’m using AI in my work, the reaction is usually skepticism. And I understand it — the idea that AI could contribute to a craft that’s thousands of years old and fundamentally about human expression feels wrong. But the AI I’m using isn’t making creative decisions. It’s handling the engineering, logistics, and planning that make large-scale installations possible.

The distinction matters. AI isn’t replacing artistic judgment. It’s taking over the computational tasks that used to require weeks of manual calculation or expensive engineering consultations.

Structural Analysis and Load Calculations

A monumental stone sculpture isn’t just art — it’s an engineering problem. A three-metre marble figure weighs several tonnes. The base must support that weight across varying conditions: wind load, seismic activity, temperature expansion, soil settlement. Getting the structural design wrong risks catastrophic failure.

Traditionally, sculptors working at this scale hire structural engineers to analyse the design and specify the base, anchoring, and internal reinforcement. That’s still necessary for final sign-off, but AI tools are now capable of preliminary structural analysis that helps me design with engineering constraints in mind from the beginning.

I’ve been using parametric design tools with AI-assisted structural analysis to test different sculptural forms against structural requirements before I commit to a design direction. I can model a proposed form, specify the stone type and dimensions, define the environmental loads for the installation site, and get a preliminary assessment of structural viability within minutes.

This doesn’t replace the structural engineer. But it means I arrive at the engineering consultation with a design that’s already structurally plausible, rather than presenting a design that the engineer then tells me won’t work. It eliminates rounds of design revision that used to stretch project timelines by weeks.

Site Analysis and Placement

Where a large sculpture sits within a space determines how it’s experienced. Sightlines, lighting patterns, foot traffic flows, neighbouring structures — all affect the visual impact of the work. For a sculpture in a gallery, these factors are relatively simple. For a sculpture in a public space, they’re complex and change throughout the day and year.

AI-powered site analysis tools can model how a proposed sculpture will interact with its environment across different conditions. How shadows fall on the sculpture at different times of day. How the sculpture reads from various approach angles. How lighting affects the perceived form and surface quality at different times of year.

I recently used this approach for a commission in a corporate forecourt. The AI analysis of sun paths showed that my initial proposed orientation would put the sculpture in deep shadow during the lunch hour — exactly when the most people would be in the space. Rotating the piece 30 degrees maintained the intended visual relationship with the building entrance while ensuring the primary face received direct light during peak foot traffic hours.

That’s not artistic judgment — it’s solar geometry. But it affected the artistic outcome by informing a placement decision that I would have gotten wrong based on a single site visit.

Material Estimation and Waste Reduction

Large-scale stone sculpture involves significant material costs. A block of Carrara marble large enough for a monumental figure can cost $20,000-$50,000 before shipping. Getting the block size wrong — too small and you can’t complete the design, too large and you’ve wasted money on stone that becomes rubble — is an expensive mistake.

AI-assisted 3D modelling helps me calculate precise block dimensions for a proposed design, including allowance for carving waste, working margins, and the inevitable adjustments that happen during carving. The models can also optimise the design’s orientation within a block to avoid known defects or grain directions that would cause problems during carving.

Teams working on custom AI development have built tools that can analyse quarry block inventories against sculptural designs, identifying available blocks that best match the requirements and minimising waste. This is particularly useful when working with expensive or rare stone types where block selection significantly affects project cost.

For one recent project, the AI analysis identified that rotating my design model 15 degrees within the block would avoid a veined section that photogrammetric quarry data had mapped. That rotation saved me from potentially hitting a structural weakness three weeks into carving — the kind of problem that used to require either starting over or making significant design compromises.

Logistical Planning

Moving multi-tonne stone sculptures from studio to installation site involves cranes, specialised transport, rigging, and careful coordination. The logistics can cost as much as the sculpture itself for complex installations.

AI tools help with logistics planning by modelling transport routes for oversized loads, calculating crane requirements based on sculpture weight and site access constraints, and scheduling the installation sequence for multi-element works.

For a six-element installation I completed last year, the AI-assisted logistics plan identified that delivering elements in a specific sequence — rather than the order I’d assumed — reduced the total crane hire time by a full day. At $3,500 per day for the crane and operator, that was meaningful savings.

Client Visualisation

This is where AI’s role is most visible to clients. Using AI-enhanced rendering, I can show clients realistic visualisations of proposed sculptures in their actual spaces. Not just a 3D model floating in a void, but the sculpture photographically composited into the real environment with accurate lighting, shadows, and material appearance.

The Unreal Engine and similar real-time rendering platforms have made this possible for several years, but AI-enhanced material rendering has improved the realism dramatically. Stone surfaces — with their translucency, grain variation, and characteristic light behaviour — are notoriously difficult to render convincingly. AI-trained material models now produce visualisations that clients find credible.

Better client visualisation leads to faster approval, fewer change requests during carving, and higher client satisfaction with the finished work. The visualisation investment typically pays for itself by reducing revision cycles.

What AI Doesn’t Do

AI doesn’t tell me what to sculpt. It doesn’t make aesthetic judgments. It doesn’t choose forms, compositions, or expressions. It doesn’t decide whether a surface should be polished or rough, whether a form should be abstract or figurative, whether a sculpture should provoke or comfort.

Those decisions are mine. They come from decades of carving experience, from studying stone sculpture across cultures and centuries, from understanding what stone can and can’t do, and from the artistic vision that makes each sculptor’s work distinctive.

AI handles the parts of large-scale sculpture that are computational rather than creative. And by handling them well, it frees me to spend more time on the work that actually matters — the carving itself.

The tools are getting better quickly. I expect that within a few years, AI-assisted planning will be standard for monumental sculpture work, the same way that CAD replaced hand drafting for architectural drawings. Not because it’s more artistic, but because it’s more practical. And practical matters when you’re trying to install six tonnes of marble in a public space without cracking it, dropping it, or blocking traffic for a week longer than planned.