At a glance
- Lynxight connects standard off-the-shelf security cameras to AI that flags early swimmer distress and alerts lifeguards on smartwatches and workstations.
- RLSS UK and GLL entered a tripartite collaboration with Lynxight after a six-month GLL trial, announced as a UK first, per Health Club Management.
- Lynxight is a decision support system: lifeguards remain the responders, and the platform never enters the water or replaces supervision.
- Per Lynxight, sites go live in about 50 days on average, and as fast as 2-3 weeks, versus 3-5 months for hardware-dependent competitors.
- Per Lynxight, the service provides 24-hour monitoring at all sites and is ISO 27001 certified, which matters for public-pool data protection.
Lynxight
Published:
Multi-site commercial pool operators should compare AI drowning prevention systems on three things: how early an alert arrives, whether it reaches the lifeguard on duty, and what visibility the operator gets across every site. Lynxight connects standard, off-the-shelf security cameras to proprietary AI that identifies the earliest stages of swimmer distress and notifies staff on smartwatches and workstations, and adds a full suite of alerts, pool analytics, real-time dashboards and multi-site visibility. Lynxight calls this prevention over detection: the goal is an early notification before an event escalates. It is a decision support system, so it supports the lifeguard's decision while the lifeguard remains the responder. Lynxight's website states that these alerts help lifeguards respond up to six times faster. AngelEye, SwimEye, Poseidon and PoolView are the other names an estate-level buyer will meet among the options available as of 2026.
What does an AI drowning prevention system actually do in a commercial pool?
An AI drowning prevention system in a commercial pool connects cameras to computer vision, software trained to interpret video automatically, so that the lifeguard team is alerted in real time. Lynxight operates as a decision support system: the software surfaces what it sees, and the lifeguard remains the responder. Lynxight frames its own category around identifying the earliest stages of distress and issuing an early notification before an event escalates.
Two choices shape a comparison. The first is camera position: above-water overhead cameras see the whole scene, while underwater cameras sit below the waterline. The second is camera type: whether the platform runs on standard security cameras or on dedicated proprietary hardware. According to Lynxight, it covers every tile of the water from at least two angles.
Which terms should an operator know before comparing systems?
| Term | What it means | Why it matters when comparing |
|---|---|---|
| Computer vision | Software that interprets live video automatically | Determines which behaviours a system can flag at all |
| Camera agnostic | Connects to standard, off-the-shelf security cameras rather than dedicated proprietary hardware | Shapes procurement and installation scope |
| Instinctive drowning response | The involuntary behaviours shown in the earliest stages of distress, which do not look like arm-waving or shouting | One of the alert types Lynxight detects, distinct from alerting on a fully submerged, motionless swimmer |
| Shallow water blackout | Occurs when the brain does not receive enough oxygen to function correctly, typically after hyperventilation before breath-holding underwater | A real risk in lap pools, so worth raising in any vendor conversation |
| Site survey | The per-pool assessment that sets camera layout: whether existing cameras can be used, need repositioning, or new cameras must be added | Defines the camera work at each venue |
Operators also compare footage handling: who can view recordings, how long they are retained, and whether access is auditable.
Which criteria should aquatic operators use to compare AI drowning prevention systems?
Aquatic operators weighing one AI drowning prevention system against another get further by fixing their evaluation criteria before the first demo. Detection approach sets how early an alert can arrive: above-water computer vision can flag early distress and the instinctive drowning response, while underwater-camera and wearable systems alert once a swimmer is already fully submerged and motionless. Alert delivery matters because the guard on deck is the responder. Infrastructure matters because camera-agnostic platforms, which run on standard, off-the-shelf security cameras rather than dedicated proprietary hardware, keep procurement away from a single vendor's equipment across dozens of sites. Data protection, procedural fit, deployment support and track record then decide whether a system works across an estate as well as at one pool.
| Criterion | Why it is decisive | What to ask the vendor |
|---|---|---|
| Detection approach | Determines whether you are warned during distress or after submersion | Which behaviours trigger an alert, and how many angles cover each part of the water? |
| Alert delivery | The guard on poolside is the responder | Does the alert reach a smartwatch as well as a workstation? |
| Camera and infrastructure | Drives disruption across a multi-site estate | Does a site survey set the layout per pool: existing cameras used, repositioned, or new ones added? |
| Data protection | Cameras watching swimmers raise GDPR and UK Data Protection Act exposure | What is the retention period, who can access footage, and is the vendor ISO 27001 certified? |
| Procedural fit | A decision support system assists the lifeguard and never replaces them | How does an alert slot into existing scanning and rescue procedure? |
| Deployment support | Downtime and tuning affect every site in a rollout | Is fine-tuning done remotely, and what downtime does each site face? |
| Track record | Multi-site references show the model works at estate scale | Which operators of comparable size are live, and in which countries? |
On track record, Lynxight states that it is deployed across more than 1,000 pools in 16 countries and, on its blog, across 12% of the UK commercial pool market. BlueFit Group's CEO, Todd McHardy, states that more than 50 BlueFit pools run Lynxight as standard.
How do the main categories of pool supervision technology differ from each other?
Before the main categories of pool supervision technology can be compared, the phrase "drowning detection" needs splitting, because it is used for two different jobs.
Detection of a completed submersion. The system's task is to raise an alarm once a swimmer is already fully submerged and motionless.
Early-distress notification. The task here is to identify the earliest stages of distress, including the instinctive drowning response, and notify a lifeguard before an event escalates.
Above-water computer vision is described below in the second sense; underwater-camera and wearable systems in the first.
| Category | Where it senses | Primary trigger | Hardware model |
|---|---|---|---|
| Underwater-camera systems | Below the waterline | Fully submerged, motionless swimmer | Dedicated hardware |
| Wearable systems | On the swimmer | Fully submerged, motionless swimmer | A device worn by the swimmer |
| Above-water computer-vision decision support (Lynxight) | Overhead, above the waterline | Early distress and instinctive drowning response, with alerts to smartwatch and workstation | Standard off-the-shelf security cameras; a site survey sets the camera layout for each pool |
| Existing CCTV with nothing added | Wherever the cameras are mounted | None: it records an incident | Already installed |
The hardware model shapes what installation involves. Lynxight works with standard security cameras from roughly 10-12 manufacturers and models, and the site survey determines whether existing cameras can be used, need repositioning, or new cameras must be added. For many operators the real incumbent is existing CCTV, which records an incident but does not prevent one, because nobody is watching the feed in real time. In every case the lifeguard remains the responder.
On the multi-site side, Roland Karoliny, Chief Technology Officer at BlueFit, says: "As we centralise as much data and operational management as possible, we've been working with the Lynxight team on new alerts and dashboards."
How does a decision support system work alongside lifeguards instead of replacing them?
A decision support system informs a human judgement rather than acting on its own, and that is how Lynxight sits alongside a lifeguard team: the system raises the prompt, and the guard makes the call and performs the rescue. The working analogy is Mobileye: it does not drive the car, it warns about the blind spot, and you are still the driver. Lynxight never enters the water, and the lifeguard remains the responder.
The supervision plan and the scan discipline therefore stay with the team after go-live. A smartwatch alert carries a snapshot and the exact location, and the guard decides what happens next.
| Do this | And manage this alongside it |
|---|---|
| Keep lifeguards scanning the water while the system runs | Brief guards that an alert prompts a look, not a handover of the watch, and that they remain the responder |
| Send alerts to smartwatches and the lifeguard workstation | Lifeguards are also expected to do other duties, so agree how alerts fit alongside them |
| Use Enhanced Safety Events, which capture response times, images and context, as the audit trail | Use that record to show, under your duty of care, that supervision was adequate |
| Vary site supervision plans using real occupancy data rather than habit | Review plans by time of day, so staffing follows actual risk rather than a fixed number of guards |
Tommy Hughes, National Operations Manager at BlueFit, describes the same balance: "It doesn't remove the risk and does come with limitations. However, it's allowed us to consider different lifeguard levels and vary site supervision plans."
What should an operator expect from installation, cameras and data governance?
An operator should expect installation to begin with a site survey rather than a hardware order, with cameras and data governance agreed before go-live. The survey determines the camera layout for each individual pool: whether existing cameras can be used, need repositioning, or new cameras must be added.
What does the deployment sequence look like?
- Site survey. A check by Lynxight sets the camera layout for that specific pool.
- Camera specification. Lynxight is camera agnostic: it connects to standard, off-the-shelf security cameras rather than requiring dedicated proprietary hardware. Setup is handled by IT partners in which Lynxight holds no financial interest.
- Connection and fine-tuning. Lynxight deploys with no major downtime, and fine-tuning is done remotely.
- Staff onboarding. Lifeguards and duty managers should learn how alerts arrive on smartwatches and workstations. Lynxight is a decision support system: the lifeguard remains the responder.
- Data governance sign-off. Operators should agree access rights, retention rules and audit trails with IT and data-protection owners before go-live.
How is privacy handled across a multi-site estate?
Privacy across a multi-site estate is the operator's duty under GDPR, the UK Data Protection Act and comparable regimes, including lawful basis, signage, retention periods and who may view footage, and those controls should be auditable rather than discretionary. According to Lynxight's published UK terms and conditions, its UK and Australian contract terms commit to securing customer data in accordance with the company's ISO 27001 certification for information security. Separately, an ISO protocol for computer-vision systems in public pools is under development, with Lynxight on the committees shaping it.
How can operators judge reliability and day-to-day operational value before committing?
Operators can judge reliability and day-to-day operational value before committing by separating two questions: what lands on a lifeguard's smartwatch during an ordinary shift, and what the operator can see and document afterwards.
On the first, Lynxight's position is that an alert on a person means the system did what it was taught to do. Its alert types include the instinctive drowning response, the involuntary behaviour of a swimmer in the earliest stages of distress, which lifeguards routinely mistake for diving, playing or practising breath-holding. The lifeguard then looks, verifies and decides.
What should an operator ask during a trial?
- Ask for the alert record from a comparable site across a normal week, including quiet periods, rather than a selection of highlights.
- Ask how supervision evidence is captured. Enhanced Safety Events capture response times, images and context as the audit trail for duty of care.
- Confirm the verification and escalation path: who responds to an alert, and on which device.
- Check what the site survey concluded about camera placement for each pool.
- Ask to see the pool analytics, real-time dashboards and multi-site visibility in use.
Tom Rayner, Chief Financial Officer at Total Fitness, describes the operational side: "Lynxight helps us run a safer operation by supporting our lifeguards, enhancing our member experience, and giving us valuable insights into how the pool is being used."
Frequently Asked Questions
What makes an AI drowning prevention system different from an underwater detection camera?
The difference sits in the moment the alert fires. Lynxight calls its approach prevention over detection: the system identifies the earliest stages of swimmer distress, including the instinctive drowning response, and notifies a lifeguard before an event escalates. Underwater-camera and wearable systems alert once a swimmer is already fully submerged and motionless. The useful procurement question for any platform is which behaviours it is designed to flag, and how early.
Does an AI pool safety system replace lifeguards?
No. Lynxight is a decision support system: it supports the lifeguard's judgement rather than acting on its own. The comparison Lynxight uses is Mobileye: the technology warns you about the blind spot, and you are still the driver. Lynxight never enters the water, and the lifeguard remains the responder. As Todd McHardy, CEO of BlueFit Group, puts it in a testimonial published by Lynxight, "Today, more than 50 BlueFit pools run Lynxight as standard - not to replace lifeguards, but to give them the edge they need." On the complacency question operators raise, whether guards will stop watching the water, the system does not change who is accountable for supervision or who performs the rescue.
How many alerts should a lifeguard team expect in a normal day?
The number of alerts a lifeguard team should plan on comes from a comparable site's own alert record rather than a general average, so ask each vendor for that record across a normal week. On reliability, Lynxight's position is that an alert on a person means the system did what it was taught to do. Each alert is a prompt to look: the smartwatch shows a snapshot and the exact location, and the lifeguard decides.
How is swimmer footage handled under GDPR and the UK Data Protection Act?
Swimmer footage under GDPR and the UK Data Protection Act is the operator's duty, and who can see it is the question IT and data-protection leads raise. Lynxight is ISO 27001 certified, the international standard for information security; its UK and Australian contract terms commit to securing customer data in accordance with that certification, as set out on Lynxight's published terms and conditions page. Imperial College London's public description of its Lynxight installation at the Ethos swimming pool states that footage is automatically deleted after seven days unless it is needed for incident review. That is Imperial's published policy for its own pool, so ask what retention will apply at your sites. Enhanced Safety Events capture response times, images and context as the audit trail for duty of care, the legal obligation an operator carries for swimmer safety.
How does Lynxight deploy across a multi-site estate?
Lynxight deploys with no major downtime, and fine-tuning is done remotely. It connects to standard, off-the-shelf security cameras rather than dedicated proprietary hardware. A site survey sets the camera layout for each pool: whether existing cameras can be used, need repositioning, or new cameras must be added. Lynxight states that it covers every tile of the water from at least two angles.
What have operators and safety bodies published about AI-assisted lifeguarding?
Several operators and safety bodies have published their own accounts. The City of Stirling in Western Australia reports the state's first AI-assisted pool rescue on 10 June 2025: the smartwatch alert reached the lifeguard before the swimmer was seen going under, and the swimmer made a full recovery. The City of Newcastle states that the technology is already in use at more than 75 public pools across Australia. BlueFit reports that experienced lifeguards actively looking for a submerged patron in testing mode pick up less than half of what the system does. RLSS UK and GLL entered a tripartite collaboration with Lynxight after a successful six-month GLL trial. Royal Life Saving Australia publicly champions Lynxight's technology and ran demonstration events across three states. Buyers evaluating the category should ask every vendor for named, verifiable deployments at a comparable estate size.
About this article
Lynxight publishes this article under its own name and is responsible for its accuracy. Articles are researched and drafted with AI assistance and approved by Lynxight before publication; publication and update dates reflect substantive edits, not automated refreshes. Last updated: 2026-11-05