A London-based convenience store has cut shoplifting losses by 90% using Artificial Intelligence (AI) technology to help tackle theft, using cameras that can tell what people are putting in their pockets.
The smart AI technology uses software from Veesion which can be connected to the existing cameras and security network through the internet. It then automatically monitors all the cameras simultaneously for potential shoplifting activity.
The tech then notifies the convenience store staff via an app. When the AI detects any suspicious gestures or movements – which can be anything from unusual browsing patterns to someone putting an item in their pocket – it sends a short video clip of suspicious activity to staff, so they can take action and deal with the suspected shoplifter.
“Once we are alerted to the suspicious activity and have dealt with it, we have three options,” explained store manager Sivakumar Pandian.
“We can either log it as ‘theft stopped’, ‘theft escaped’ or ‘no further action’ – which tells the AI system that it wasn’t a suspicious activity after all. This helps it learn what needs to be reported and what is normal shopping activity.”
Over time, as staff use the app to confirm or disregard any potentially suspicious activity, the AI will be able to teach itself the difference between a customer putting their mobile phone away and slipping a product from the shelf into their pocket.
The convenience store, a Nisa located in Virginia Quay, was originally losing up to £1,000 a week as consumers helped themselves to products without paying for them. Losses are now less than £100 per week.
“Shoplifting had always been bad here, but it became a real problem over the past three or four years,” said store manager Sivakumar Pandian.
“Shoplifters were now starting to fight back and staff were getting injured. It’s not acceptable and so we had to think about new ways to address the problem.”
Initially, Pandian added a new camera system and bought in additional security detail. But it wasn’t enough, especially during Covid, when staffing issues made thefts even harder to monitor.
“We needed something that allowed us to be everywhere at once, even when staff were thin on the ground,” Pandian said.
“The new system allows us to do exactly that. It was effective from the very first day we installed it and the technology is only going to get better as time goes on.”