QuantumSpace Uses Knowledge Graphs to Combat Art Fraud

QuantumSpace, an AI start-up specializing in computer vision and visual intelligence, is leveraging knowledge graph technology to address art fraud. The company's system represents thousands of artworks as a knowledge graph with billions of nodes, detailing relationships between them. Francesco Rocchi, COO, believes this can transform subjective authentication into data-driven analysis. The technology has already identified an undocumented restoration in a painting attributed to Caravaggio by flagging inconsistencies in brushstroke analysis. It also aids in inspecting craquelure, the fine cracks on aging paintings. The global fine-art market, estimated at $60 billion in 2025, faces forgery losses up to $1.6 billion annually. QuantumSpace, founded in Italy in 2019 and headquartered in New York with an office in Milan, uses Neo4j's graph database. It has archived 68,000 paintings, generating 3.5 billion data points. Milan gallery owner Deodato Salafia sees potential in customer engagement and providing structured technical information to professionals. Rocchi's background in psychology and visual arts inspired the idea of mapping an artist's cognitive process from artwork data. The company collects up to 60,000 data points per painting, spanning over 50 datasets including color, surface physics, condition, and provenance. The technology is also applied in biomedical imaging and automotive inspection.