On May 1 at the Waterlane Leisure Centre in Lowestoft, Suffolk, a five-year-old boy nearly drowned after an AI-powered drowning detection system malfunctioned because a torpedo buoy was left on its sensor button. An on-duty lifeguard rescued the unresponsive boy, who was underwater for about three minutes, and he made a full recovery after receiving CPR and hospital care. five lifeguards have since been dismissed, including two who were on duty at the time and a teenager who was not present during the incident, according to Everyone Active, the leisure centre operator.

The torpedo buoy that blocked a life-saving alert

The system at Waterlane uses eight underwater cameras and sensors linked to an AI monitoring system designed to track swimming activity and send real-time alerts to lifeguards. However, as the source reports, a torpedo buoy is understood to have been placed on the sensor button that emits the alert, effectively silencing the system. A member of the public spotted the boy underwater just after a swimming lesson, prompting the rescue that the AI should have flagged earlier.

Stuart Jardine, area contract manager for Everyone Active, confirmed the rescue and said the boy was taken to hospital as a precaution. The incident has raised serious questions about the reliability of automated surveillance systems when physical objects can disrupt their sensors.

Two on-duty lifeguards sacked — and a teenager who wasn't there

According to Everyone Active and East Suffolk Council, five full-time staff were dismissed after disciplinary hearings and investigations. Among them is a teenager from Lowestoft who was working earlier that day but had already left before the incident. In a copy of his dismissal letter seen by a local newspaper, the reason cited included “breaches of critical health and safety practices including lifeguarding malpractice and the misuse of the drowning detection system.” The teenager, who is appealing the decision, told the source: “It is galling, especially as I was only working during the day and was not even there in the evening. I did not think anything of it as I was new back then.”

The inclusion of a staff member uninvolved in the actual rescue has drawn criticism, with many asking whether the dismissals are proportionate or if they reflect a desire to place blame quickly.

What the AI system promises vs. what happened

AI-powered drowning detection is marketed as an extra layer of safety that can spot distress faster than human eyes. Systems like the one at Waterlane are supposed to alert lifeguards within seconds of irregular movement or submersion.. Yet the source reveals that a simple buoy left on a sensor button was enough to defeat the system entirely. The boy was underwater for three minutes before a patron spotted him — a gap that undermines confidence in the technology’s real-world reliability.

This incident echoes broader debates about automation in safety-critical roles. while the AI was meant to assist lifeguards , its failure placed the entire burden back on human vigilance — which fortunately still worked, but only after a bystander intervened.

Ongoing investigation: what remains unknown about the buoy and the dismissals

East Suffolk Council has stated that the investigation is likely to be lengthy, and has declined further comment until it is complete. several key questions remain unresolved: Exactly who placed the buoy on the sensor? Were there any prior warnings about the system being blocked? And why was a teenager who was not on the pool deck during the near-drowning held accountable for failing to report an issue he said he only witnessed once as a new employee?

The source notes that the teenager claims his failure to report was due to being new, and he believes health and safety protocols should have flagged the practice earlier. without clear answers, the dismissals risk appearing arbitrary, especially as the boy’s own survival depended on human action rather than the AI.