How long does it take to find a hand played two hours ago?
A dispute stops being six hours across four cameras.
You describe the hand, IRIS finds the moment and compliance reviews the facts. We show twelve cameras here: nine that raise an alert when something departs from what was written down, and three that alert nobody, they only measure. A gaming floor has hundreds. The system is the same. And one thing before you go in: IRIS identifies nobody.
Without IRIS this is just a busy floor on a screen.
AI-generated image
A floor understood, without identifying anyone
A player raises a dispute. Somebody goes down to the recording room, picks four cameras and spends the afternoon rewinding. Meanwhile, upstairs there is a table with people standing and waiting while another one sits open and empty five metres away, a machine has been jammed for half an hour and nobody knows, and the queue at the cage has eaten the good mood of the whole floor. All of it happens in front of cameras that are switched on.
IRIS watches for them. It analyses every camera at once and only interrupts when a rule the floor itself wrote is met: a table that stops, a machine nobody has played for too long, a cage queue that keeps growing, a cash move that does not follow the usual route. And for everything else, describing it is enough: you ask for the hand by what happened and it is in front of you. What IRIS does not do: it identifies nobody, it recognises no faces, it carries no biometrics, it judges nothing and it concludes nothing. It sees situations, timestamps them and keeps the video.
What arrives is not more video. It is an alert with the camera, the time and what is happening, and the minute of recording where it started. And when nothing happens, it is numbers: how many people are at each table, how long people stand waiting for a seat, how long a hand lasts, how long the cage takes. Two limits, stated here and not in a footnote: the money a table or a machine makes comes from the gaming system, not from the camera; and a dealer sequence flagged for review is exactly that, a sequence for compliance to look at, never an accusation against a worker. Same with the movement of value across the floor: the camera contributes a dated fact, and the person who decides is an analyst, cross-checking it against the transactions.
The real scale
Twelve here. Hundreds in a real gaming floor.
This demo shows twelve cases because twelve fit on a screen. A real gaming floor has a camera over every table, another over each bank of machines, and more at the cage, in the count room, in the service corridors, at the entrances, at the cloakroom and all along the perimeter. Hundreds of them, and a few more with every refit. No person can watch them all on a Saturday night; that is exactly the problem IRIS solves: it analyses however many there are, all at once, and only interrupts when a rule is met. And where there is nothing to watch for, it measures: occupancy per table and per time slot, the wait to get a seat, the time between hands, the queue at the cage. That is the data you use to decide which table opens on Friday, and at what time.
Examples
These twelve are just examples.
Each one shows a different kind of thing: a table, a machine, a queue, a dispute, a cash move, a fall. They are not the twelve things IRIS can do: they are the ones that fit on a screen. Open any of them and you will see the image as it is, and then with what IRIS understands on top.
Demo · IRIS at work
No extra alarms here. Just what needs looking at.
Not a recorded video, not an explanatory drawing: a summary of the real screen, with a different case every time it starts. There is a lot more behind it.
- You ask
- IRIS searches
- It shows you
Now think about yours
Imagine finding the exact minute without going down to the recording room.
A table with people standing and waiting while another sits open and empty five metres away. How long people stand before they get a seat. A table that has lost its rhythm, and a dealer away from their post. A machine that has been down for half an hour and nobody has seen it. A cage queue with two windows closed. A dispute under way, on foot, in front of the table. A chip that moves after betting has closed. A dealer sequence flagged for compliance to look at. A run of drop boxes that takes a different route. The count room with two people inside, as the procedure requires. How often the cage is visited, and how far apart. And somebody on the floor while, at the same time, the door of a technical cabinet is being tampered with.
All of that is happening in front of cameras your floor already owns and already paid for. IRIS watches them all at once and tells you while it is happening, with the camera and the table. And there is a line it does not cross: minors and self-exclusion are only handled where there is a legal basis and an authorised identification system, and that system is not IRIS, because IRIS carries no biometrics. Without identifying anyone, what it can measure is how long a play session lasts, a fight, a fall, an entry into a restricted area and the tampering with a camera. IRIS contributes facts and video; the compliance is signed by whoever runs the floor.
Next step
What does a shopping centre see beyond the visitor count?
12 cases: 4 raise an alert when something happens and 8 only measure. You see each image as it is, and then with what IRIS understands on top.
What does an airport see before the queue forms?
14 cases: 7 raise an alert when something happens and 7 only measure. You see each image as it is, and then with what IRIS understands on top.
And the best part
Sometimes you do not need an alarm:you need a number.
How many people came in, and when. You draw a line over the image and the camera counts. People counting is certified by the Centro Español de Metrología through NeuralPax.
Tell us your problem and we will say whether IRIS understands it, or not yet.
We reply the same working day.