A documented comparison for security teams that want unusual-event detection across a large camera count. icetana names its detections, which is more than several better-known vendors manage. What it does not publish is how the platform gets onto your cameras, or what it costs.
icetana sells AI security video analytics built around detecting unusual events and behaviour, with named alerts for loitering, fire and smoke, aggressive behaviour and people, bikes or vehicles in unusual areas, aimed at cutting the false alarms that make guard monitoring unreliable. Teams shop for an alternative when they need camera compatibility and deployment answered in writing, when the same cameras have to cover safety and operations too, or when something has to happen on site after the alert. The five below answer those differently.
Both platforms promise fewer, better alerts from the cameras a site already has. They differ in how much of the decision you can make from published documentation.
Every cell is something the vendor publishes, or an explicit not publicly specified.
Swipe the table sideways to see every column.
icetana's row was read off icetana.com on August 12, 2026. The named detections are its own published list; the four blank cells are absences on its public pages rather than gaps in this research, and they are the questions to put to the vendor first. The other five rows come from each vendor's public documentation reviewed in July and August 2026 on the same five columns.
Start by discarding the wrong reason, and give icetana credit for the thing this category usually fudges. It names its detections. Loitering, fire and smoke, aggressive behaviour, and people, bikes or vehicles in unusual areas are all published, along with the framing that matters most to anyone who has run a guard room: fewer false alarms, because a real-time detection that cries wolf is worse than no detection at all. Several larger vendors on this page publish nothing that specific.
The first real reason is everything the pages do not say. There is no camera compatibility statement, no protocol, no conformance profile, and no deployment model. For a platform whose entire premise is running on the cameras a site already has, that is the first question a buyer needs answered, and it is currently a conversation rather than a document. Ask for the stream requirements in writing and check them against your oldest recorder before this gets onto a shortlist.
The second is scope and what follows the alert. The published set is security shaped: unusual events, aggression, loitering, fire. It says nothing about whether the dock was blocked for forty minutes, whether the crew wore harnesses, or how long the queue ran at shift change, and no speaker, talk down or strobe behaviour appears. So the escalation path ends with a person looking at a screen, which is fine on a staffed site and is the whole gap on an empty one.
A named detection list is worth more than a page of adjectives. It is worth less than a named detection list plus an answer on cameras, deployment and cost.
What each one is, where it is strong, and what to check before you commit.
Spot AI is a software-led video AI platform that turns the IP cameras a business already owns into AI coworkers. An on-site Intelligent Video Recorder keeps full-resolution video in the building and sends only event metadata to the cloud, so search and multi-site management stay cloud-based while footage does not leave the facility.
Eagle Eye Networks is a cloud video management system that connects to virtually any ONVIF-conformant camera and digitises analog feeds through analog-ready Bridges and CMVRs, then layers cloud analytics on top of the estate. Its documentation names ONVIF Profile S and dual codec streaming as the integration path.
Coram AI is a camera-agnostic platform built around AI search and investigation across cameras a business already owns, with named detections for firearms, falls and PPE violations, and alerts that can be built in plain English. An on-site appliance called Coram Point is purchased upfront.
Verkada is a cloud-managed platform that sells its own cameras, access control, sensors, alarms and intercoms under one console, and publishes per-device MSRPs, which is unusual in this market. Third-party cameras come in through Command Connector, which Verkada states is an ONVIF Profile S conformant client and can also ingest RTSP feeds.
Rhombus sells cloud-managed cameras, sensors, access control and alarm monitoring as one system, described by the company as a cloud-edge system that operates offline. Third-party cameras come in through the Relay Connector N100, and the company publishes SOC 2, NDAA, GDPR and TAA compliance badges.
Six questions that separate these platforms faster than any feature list.
icetana publishes no camera compatibility statement, so this is question one. Eagle Eye publishes more than 7,500 models plus analog through Bridges. Spot AI takes any ONVIF or RTSP camera and legacy analog through the Intelligent Video Recorder. Coram says any IP camera without naming a protocol. Run every answer against the oldest camera on your list.
No deployment model appears on icetana's public pages: cloud, on premises and edge are all unstated. Eagle Eye stores video on its own data centers. Spot AI keeps full-resolution video in the building and sends only metadata. On a segmented or air-gapped network this decides whether a pilot is even possible, so ask before you demo.
Reduced false alarms is icetana's central promise and the right one for a guard room, but no figure or method is published. Turn it into a pilot specification: which sites, which alert types, over how many weeks, against what baseline, and what counts as a false positive. Any vendor selling on this should be comfortable defining it.
Ask what the anomaly triggers at 02:00 beyond an entry in a list. Eagle Eye documents sirens and talk-down alerts. Verkada gates automated audio to the BZ11 or an Intercom. Spot AI runs talk down, strobes and horns through standard speakers already mounted. icetana, Coram and Rhombus publish no deterrence behaviour, so the honest answer is to ask.
If your list includes PPE compliance, forklift near-misses, hazard-zone crowding or dock delays, ask which of those ship pre-trained. icetana's published set is security shaped, and none of the catalog vendors here publishes an operations or safety agent set, so a platform that arrives with them answers a question the others route elsewhere.
icetana publishes nothing on cost. Verkada publishes per-device MSRPs. Coram publishes a licensing model without figures. Everyone else here, Spot AI included, publishes nothing. Insist on the same site list from every vendor, itemised into software, hardware, installation and storage, at one site and at your full camera count.
Spot AI fits when the cameras are already mounted, the questions span security, safety and operations, and the platform is expected to act rather than notify. It runs on any ONVIF or RTSP camera plus legacy analog through the Intelligent Video Recorder, keeps full-resolution video on site, and ships 15+ pre-trained agents plus Iris for custom detections.
icetana stays a reasonable call for a security operation with a large camera count whose specific pain is unusual events getting lost in noise, and which is happy to settle camera compatibility, deployment and cost in conversation rather than off a website. The named detection list is a genuine strength and the blanks are the work.
A camera-agnostic platform lets you test the AI on your own cameras before any hardware decision, which is the cheapest way to de-risk the choice.
Watch the AI coworkers work on your live feeds, not a demo reel.
Customer-reported outcomes from named Spot AI customers.
Cambridge City cut the time it takes to find footage from two hours to thirty seconds.
Storage Asset Management was alerted to a break-in around one in the morning, and law enforcement arrived while the incident was still in progress.
Bridge33 Capital standardized video across 25 plus commercial properties on whatever cameras each acquisition already had, and cut footage search from hours to minutes.
"The Spot AI notification went out around one in the morning, and police were able to arrive on-site while the incident was still in progress."
The five covered here are Spot AI, Eagle Eye Networks, Coram AI, Verkada and Rhombus. Spot AI fits teams wanting security, safety and operations from the same cameras with deterrence on top. Eagle Eye Networks fits mixed and partly analog estates with documented camera support. Coram AI fits investigation-first teams. Verkada fits a site specified from scratch. Rhombus fits mid-market teams wanting one vendor.
Its pages do not say. No protocol, conformance profile or compatibility list appears publicly, which for a software analytics platform is the first thing to establish. Ask for the stream requirements in writing, then test them against your oldest cameras rather than your newest, because that is the camera that decides the project.
Its published alerts include loitering, fire and smoke, aggressive behaviour, and people, bikes or vehicles in unusual areas, all framed around real-time detection of unusual events with fewer false alarms. No safety or operations detections, and no deterrence behaviour, appear on the public pages.
No. No price, plan tier or contract term appears on its site, which is the norm in this category rather than an outlier. Of the platforms compared here, only Verkada publishes per-device MSRPs and Coram AI publishes a licensing model without figures. Any comparison has to come from quotes on the same camera list.
That is icetana's own pitch, so test it directly and test the alternatives on the same terms. Ask each vendor how detections are tuned per site, what the review workflow looks like, and whether an alert can trigger an action rather than a notification. Spot AI's agents are pre-trained and tunable, and its deterrence runs on site, which changes what a confirmed alert is worth.