Home / Crime and Justice / AI Unmasks Art Thieves: A New Era of Detection
AI Unmasks Art Thieves: A New Era of Detection
28 Oct
Summary
- AI system detected a disguised 17th-century painting stolen over 20 years ago.
- Italian AI has recovered over 300 paintings and 165,000 smaller items worth €203m.
- AI identifies stolen art by matching visual fingerprints, even with disguises.
Italian police are enhancing their fight against art theft with a novel AI software, Swoads, which significantly sharpens their investigative capabilities. This AI system excels at identifying stolen artworks by matching hidden visual fingerprints, even when pieces have been altered or disguised. For example, it recently flagged a 17th-century painting stolen more than 20 years prior, which had been repainted to conceal its identity.
Since its deployment in November 2023, Swoads has been instrumental in recovering over 300 large plundered artworks and more than 165,000 smaller items, with a total value exceeding €203 million. The AI continuously scans online marketplaces and databases, making art recovery faster and more efficient than traditional manual searches.
While AI is a powerful tool, experts note it's a "force multiplier" rather than a complete solution. It aids in identifying stolen items but does not apprehend thieves or recover pieces never recorded. Challenges remain, including refining AI's ability to recognize 3D objects and ensuring sufficient personnel to follow AI-generated leads.
Globally, other countries are also developing AI for art theft investigations, with initiatives in Turkey, France, Germany, and the US. Interpol's ID-Art software connects databases internationally. Despite AI's advancements, recovering highly prized or recently stolen works remains difficult, as these are often hidden by collectors. However, AI remains invaluable for identifying lesser-known pieces and restoring cultural heritage, as demonstrated by the return of a 17th-century painting to a small church in Naples.
