A structured review of Oosto's identity-led Vision AI platform for security teams whose requirement is recognising named individuals: what it does well, where the constraints sit and who should look elsewhere. Every score traces to something Oosto publishes.
Oosto is built around a question most platforms in this category do not attempt: who is that, right now, in a crowd. It publishes many-to-many face recognition, person of interest alerting, liveness detection using a 3D face map against spoofing and cross-camera tracking, across three named products for cloud alerting, watchlist monitoring and access control. It is 100% VMS agnostic in its own words, with named integration for Honeywell, Milestone and Genetec, and it publishes its privacy controls: blurring non-watchlisted individuals on playback, images converted to vectors including at the edge, and a watchlist the customer builds and maintains. It scores lower on everything outside identity, because the behaviour set is security shaped and no safety or operations agents appear, and lower on deployment, because where video is retained and for how long is not publicly specified. The capability is the straightforward half of this decision. Running it in Illinois, Texas or Washington means running a programme under a dedicated biometric statute, and the Illinois Act carries a private right of action. Choose Oosto when recognising specific people is a genuine requirement and the organisation can carry that. Look elsewhere when the question is what happened rather than who it was.
Five criteria, each scored from Oosto's own current documentation as of August 24, 2026. Spot AI competes with Oosto, so no score here rests on an anonymous source, an aggregated user rating or a private benchmark: each one sits next to the documented fact behind it. Anything Oosto does not document is recorded as not publicly specified rather than assumed. This is a privacy-sensitive category, so this page reports what Oosto publishes and names the statutes a buyer runs under rather than judging the technology. This is not a paid placement, and Oosto had no input into it.
The identity path is documented end to end, including the privacy controls along it. Where the video sits and what fires after a match are the two that are not.
Oosto sells Vision AI organised around identity. Three products are named: Oosto Protect for edge-to-cloud alerting on face, body and behaviour, Oosto OnWatch for watchlist monitoring with live face recognition, and Oosto OnAccess for opening guarded points of entry for authorised people using the cameras already installed. The technology pages carry the specifics: many-to-many face recognition, simultaneous recognition of individuals in crowds within a fraction of a second, and liveness detection that builds a 3D face map to catch spoofing attempts.
Compatibility is stated as broadly as any vendor in this review. Oosto describes itself as 100% VMS agnostic plus integration with Honeywell, Milestone and Genetec, and publishes an edge-optimised design that pushes more processing onto cameras through embedded SDKs and onto near-edge devices. Deployment is via cloud, on premises or SDK, with published timelines of as little as one week on premises and under an hour in the cloud.
The privacy controls are published rather than left to a policy document. Oosto states that non-watchlisted individuals can be blurred on video playback, that image data is converted to vectors including at the edge, and that the customer builds and maintains their own watchlist with Oosto never seeing that data. It publishes material on GDPR and on the EU Artificial Intelligence Act, and names its estate types openly, from airports and seaports through casinos, critical infrastructure and healthcare to retail, schools and stadiums.
Oosto is strongest when the requirement is genuinely who, and thinnest the moment the question becomes what happened and what should have fired.
Both columns are documented. Nothing here comes from an anonymous review.
Both columns describe documented behavior. Oosto's column was checked against its own documentation on August 24, 2026.
Swipe the table sideways to see every column.
Oosto data comes from Oosto's own public documentation as checked on August 24, 2026. Gaps are marked as not publicly specified.
The honest split, stated the way a shortlist call would state it.
If the job is knowing that a specific person is at a specific door or in a specific crowd, and your organisation has the legal basis and the governance to carry it in Illinois, Texas and Washington, Oosto is built for exactly that and publishes it in detail. The VMS-agnostic design, the named integrations and the published privacy controls make it addable to an existing estate rather than a replacement for one.
If the questions are a near miss on the floor, dwell at a dock door or what should have fired at 03:00 on an unstaffed site, identity is the wrong lens and the published set does not reach there. Deterrence behaviour is not publicly specified either. A platform that never builds a face template also never triggers the Illinois, Texas or Washington biometric statutes, which is a compliance cost avoided rather than managed.
Spot AI ships 15+ pre-trained Video AI Agents spanning vehicle break-in, fire, intrusion, personal protective equipment, forklift near-miss, falls and hazard-zone crowding, so a mixed estate gets shipped coverage on both sides. Iris builds anything else in natural conversation in about eight minutes, full-resolution video stays on the Intelligent Video Recorder in the building and the platform is SOC 2 Type II, NDAA-compliant and HIPAA-aligned.
The first thing to settle is whether these two are alternatives at all, because on many shortlists they are not. Oosto's published capability is identity led and Spot AI's published agent set detects events and behaviours rather than matching a person against a watchlist. If real-time recognition of named individuals is a genuine requirement, that decides it before any other row is read.
If the requirement turned out to be behavioural rather than identity, run the comparison properly: one site, a fortnight, both platforms live on the same cameras, and a site whose incidents are behavioural rather than identity driven. That is the version of this comparison where the answer is genuinely in doubt.
Put the compliance work on the same sheet as the licence. A biometric programme carries legal basis, retention policy, consultation and a template deletion plan that a non-identity detection set does not.
See what the AI catches on your live feeds before any platform decision.
Customer-reported outcomes from named Spot AI customers.
Liberty-Perry School District resolves an incident in about five minutes, after evaluating ten systems before choosing Spot AI.
Cambridge City cut footage search from two hours to 30 seconds after consolidating seven municipal locations on one platform.
Unique Industries covers more than a million square feet with a three-person safety team, catching near misses and falls on the cameras already installed.
"The biggest benefit of Spot AI is how easy it is to look up incidents, see the footage, and then share it and collaborate."
Yes, for the requirement it is built for. It scores 3.6 out of 5 here, carried by depth on identity, many-to-many face recognition, recognition in crowds and liveness detection with a 3D face map, by 100% VMS agnostic compatibility with named integrations and by published privacy controls. It fits security teams whose genuine requirement is recognising specific individuals, with the governance to run that. It fits poorly where the question is behavioural or operational.
Yes. Oosto describes itself as 100% VMS agnostic plus integration with Honeywell, Milestone and Genetec, and publishes an edge-optimised design that pushes processing onto cameras through embedded SDKs and onto near-edge devices. A camera protocol or conformance profile is not publicly specified, so an unusual fleet is worth confirming directly before a rollout.
Several, and they are published rather than left to a policy document. Non-watchlisted individuals can be blurred on video playback, image data is converted to vectors including at the edge, and Oosto states the customer builds and maintains their own watchlist and that Oosto never sees that data. GDPR and EU Artificial Intelligence Act material sits alongside it. What is not publicly specified is where full-resolution video is retained and for how long.
Oosto published the acquisition itself, announced on 22 January 2025, and stated that its team and employees would join Metropolis while providing continued support and service to all current and prospective customers and partners. oosto.com now carries an "Oosto is now Metropolis" banner pointing at metropolis.io/security, so the product pages are still up while the brand has moved. Ask which line a renewal renews rather than reading either page as an answer.
It depends on whether identity is the real requirement. If it is, nothing on a shortlist replaces it and that line should be kept. If the estate needs to know what happened and to have something fire when it does, a platform such as Spot AI fits: 15+ pre-trained Video AI Agents span security, safety and operations on the cameras already installed, full-resolution video stays on the Intelligent Video Recorder, and an incident is answered with talk down, strobes and horns through standard speakers. It also builds no face template, so the biometric statutes do not apply. The alternatives roundup compares five platforms side by side.