🛡️ AI Detects Shoplifting
Paris-based startup Veesion has developed a gesture-recognition algorithm that can predict theft in retail stores. According to the company, its technology is already used in 5,000 stores across Europe, Canada, and the U.S. Veesion recently raised €53 million to accelerate its expansion in the American market.
"My uncle runs three supermarkets in Paris, so I know exactly what shoplifting means for retailers," says Veesion co-founder Benoît Koenig.
❔ How It Works
The system uses footage from regular surveillance cameras to detect people, track their movements, and identify objects, such as products, carts, baskets, and bags. If a movement looks suspicious, the system flags the footage and alerts store security.
According to Koenig, the alerts are accurate in about 85% of cases. One client reportedly cut theft in half within the first three months of using the system.
Veesion emphasizes that the system does not track appearance or use biometric data, reducing concerns about bias or privacy.
It also helps reduce internal theft from employees.
"It has a deterrent effect. People know there's AI watching, so they're more careful," Koenig explains.
➡️ What's Next?
Veesion is expanding the use of its technology beyond theft detection to include tools for inventory control and customer behavior analysis. In the future, the company sees applications beyond retail: in hospitals, warehouses, and construction sites, where motion analysis can improve safety.
