Repeat shoplifter gets caught and jailed after being identified by Sheng Siong's AI facial recognition security system - Singapore News

SINGAPORE: A 51-year-old woman was jailed for eight days after Sheng Siong’s facial recognition security system flagged her return to a supermarket outlet.

The woman, Catherine Tan Li Eng, pleaded guilty to stealing 19 bottles of Jacob’s Creek wine worth about S$556 over seven visits to a Sheng Siong outlet in Punggol in September 2025. The case has also reopened conversations about how retailers are increasingly relying on artificial intelligence (AI) tools to address repeat theft in stores across Singapore.

Sheng Siong staff discovered stock discrepancies before reviewing closed-circuit television (CCTV) footage. The footage allegedly showed Tan placing bottles of wine into her bags before scanning only lower-value items at the self-checkout kiosk, according to Channel NewsAsia (CNA), May 12 report.

The AI alert led staff straight to the self-checkout kiosk before she could leave again

After reviewing the footage, staff uploaded Tan’s image into Sheng Siong’s AI-powered facial recognition system. The following day, the supermarket’s retail manager received a mobile alert after Tan entered the store again. He approached her at the self-checkout area and checked her bags after she claimed all items had been paid for.

Three bottles of wine were allegedly found unscanned inside her bags. Police were then called, and the bottles were recovered. Tan also repaid about S$470 for the remaining stolen items.

The case gives a closer look at how supermarkets are now using technology beyond traditional CCTV systems. Instead of relying solely on staff memory or manual monitoring, facial recognition software now flags repeat suspects almost instantly upon their entering a store.

Retailers are now turning to AI facial recognition security systems to deter shoplifting

Sheng Siong first rolled out its AI-driven facial recognition CCTV system in April 2024 before expanding it to all outlets in August 2025 to curb shoplifting. At the time, shop theft had become one of the more common offences involving young people in Singapore.

Retail theft remains a difficult problem for supermarkets because losses frequently happen in small amounts across repeated visits. Self-checkout kiosks have also made theft detection harder, especially during busy periods when staff are monitoring multiple counters at once.

The latest case shows how supermarkets are going toward automated monitoring systems that can react faster than human staff alone.

While some Singaporeans may welcome tighter enforcement against repeat shoplifters, others are likely to pay closer attention to how facial recognition systems are used in everyday retail spaces.

The court says her repeated theft raised the seriousness of the case

Defence lawyer Mr T M Sinnadurai said Tan was the sole caregiver for her mother and intellectually disabled cousin. He asked the court for a shorter jail term and said she regretted the offences.

District Judge Brenda Chua rejected the request for a one- to two-day sentence, saying the amount stolen was substantial and the repeated offences increased the seriousness of the case. And because the charge combined several theft incidents into one amalgamated charge, Tan faced tougher penalties under the law.

The case may sound like a straightforward shoplifting story, but it does show a serious change underway across Singapore’s retail sector. Artificial intelligence is no longer just a future concept hiding in the background. It has, at present, become part of daily enforcement in supermarkets, watching shelves, tracking behaviour, and alerting staff in real time.

For shoppers, the simplest solution remains the oldest one: pay for what goes into the basket. Technology may be getting smarter, making it easier for shoplifters to take advantage of or even abuse it, but retail theft always leaves a very human trail behind for the very technology to identify them back.

- Advertisement -