Vaidio, formerly known as IronYun, provides an AI video-analytics platform for security monitoring, access control, forensic investigation, operational analysis, and selected industrial applications. The software adds analytics to compatible cameras and video-management infrastructure, allowing organizations to configure event detection, search recorded footage, track objects across supported views, and generate alerts.
Organizations evaluating video analytics should define their required use cases, camera conditions, response workflows, privacy requirements, and system integrations before selecting a platform. NIST’s guidance on interoperable video analytics highlights governance and information-system interoperability as important considerations when analytics are introduced across complex camera environments.
IronYun Inc. USA was incorporated in 2015 and established its headquarters in Stamford, Connecticut. In March 2025, the company adopted Vaidio as its primary brand alongside the release of the ninth generation of its AI Vision Platform.
IronYun remains the legal entity identified in the company’s current software agreement, while Vaidio is the name used for its platform, website, product releases, and market positioning.
Vaidio 10.0 became generally available on July 8, 2026. The release added recorded-video analysis, support for newer NVIDIA architectures, changes intended to improve GPU utilization and deployment scale, and VLM-assisted verification through Vaidio Vista.
Vaidio is designed for organizations that want to apply several forms of video analytics across existing camera environments. Its published applications span:
This breadth allows security and operations teams to apply different analytics through one platform. It also means buyers need to identify which functions are relevant to their actual workflows rather than evaluating the product only by the total number of available analytics.
Vaidio supports configured alerts for common site-security and surveillance scenarios, including:
These capabilities can help operators review events across camera networks that are too large to monitor continuously through manual observation alone.
Performance will depend on factors such as camera placement, image quality, lighting, weather, object size, scene complexity, thresholds, and the selected analytic. Buyers should test each required scenario under normal operating conditions.
Vaidio indexes video metadata and allows users to locate people, vehicles, objects, and activities in recorded footage.
Search functions include:
Vaidio advertises sub-two-second forensic searches under supported conditions. Actual performance will depend on the number of indexed cameras, stored footage, hardware, architecture, network performance, and query complexity.
Vaidio supports identity and access-control applications such as:
These capabilities may be useful where security teams need to connect video information with existing access-control systems.
Facial recognition and related identity analytics introduce legal, privacy, workforce, consent, and accuracy considerations. Organizations should confirm whether these functions are lawful and appropriate for each location and establish a human-review process before acting on a match.
Vaidio publishes manufacturing and industrial applications involving security, safety, quality, logistics, and operations.
Examples include:
The platform’s broad library may allow one camera environment to support several departments. It does not establish that every detector has the same accuracy, maturity, latency, or suitability for safety-critical decisions.
Industrial buyers should request the current detector specifications, camera requirements, alert timing, validation data, and references for each proposed use case.
Vaidio supports custom analytics for objects, activities, or conditions that are not covered by its standard models. Customers can provide labeled footage and work with the vendor to develop a site-specific application.
Potential custom use cases include:
Custom development may require additional video collection, labeling, training, validation, tuning, and maintenance. Buyers should clarify model ownership, update procedures, retraining requirements, and support when site conditions change.
Vaidio works with many ONVIF-compatible IP cameras and integrates with video-management and recording systems from providers such as Milestone, Genetec, Axis, Bosch, Avigilon, Digital Watchdog, Hanwha, and Network Optix.
Deployment options include:
This flexibility can accommodate different data-location, latency, network, and cybersecurity requirements. Hardware needs depend on camera count, resolution, frame rate, analytics, retention, search activity, and the number of models running concurrently.
Vaidio states that initial software installation can be completed quickly when compatible hardware, cameras, network details, and system access are already available. A complete deployment may take longer because it can also involve server procurement, network configuration, VMS integration, privacy review, alert design, operator training, and acceptance testing.
Vaidio combines security, identity, investigation, operations, traffic, and selected industrial analytics within one platform.
This may be useful for organizations that want to avoid operating separate systems for perimeter alerts, license plates, forensic search, crowd analysis, facial recognition, and other security functions.
The practical value depends on how many of those functions the organization will implement and maintain. Buyers should define the required use cases before determining whether a broad platform is preferable to a narrower application.
Support for ONVIF-compatible cameras and multiple VMS platforms may allow organizations to retain much of their current surveillance infrastructure.
Existing cameras still need to meet the requirements of the selected analytics. A camera suitable for general surveillance may not provide the resolution, angle, frame rate, or object visibility needed for automated PPE, facial, vehicle, or safety detection.
The combination of metadata indexing, cross-camera tracking, recorded-video analysis, and natural-language search may reduce the time required to locate relevant footage.
These capabilities can support:
Organizations still need policies for footage access, evidence export, retention, chain of custody, and authorized use.
On-premises, edge, cloud, and hybrid models give security and IT teams options for balancing latency, bandwidth, storage, infrastructure, and data-residency requirements.
Vaidio also announced SOC 2 Type II attestation in June 2026. This provides independent review of defined security controls over a period of time, but it does not replace the customer’s own assessment of architecture, authentication, logging, encryption, subprocessors, incident response, and deletion processes.
The current G2 reviews show a 4.8 rating based on 18 reviews. Reviewers frequently mention integration flexibility, feature breadth, search functions, ease of use after configuration, and responsive support.
Reported limitations include:
The sample remains small, and several reviewers work for resellers, integrators, consultants, or security-technology providers. Buyers should combine public reviews with references from end users operating comparable sites.
A broad analytics platform requires configuration across cameras, models, thresholds, alerts, permissions, hardware, and operator workflows.
Implementation may involve teams from:
Without defined ownership, the system may generate alerts without a consistent review or response process.
Vaidio publishes industrial applications involving PPE, falls, hazardous areas, fires, leaks, and vehicle-pedestrian events. Public materials do not provide detailed sensitivity, false-positive, or missed-event results for every industrial detector.
Testing should reflect:
A platform-wide accuracy statement should not replace validation of the exact analytic and environment under consideration.
Vaidio’s current terms advise customers not to rely on the product for safety-critical decisions without appropriate human review and safeguards.
Video analytics may support awareness, prioritization, investigation, and intervention. They should not replace:
The operating procedure should distinguish informational alerts from conditions requiring immediate action.
Facial recognition, appearance search, and cross-camera tracking can support security investigations, but they also create additional governance requirements.
Organizations should review:
Privacy review should cover the complete environment, including cameras, recorders, VMS platforms, analytics servers, cloud services, networks, and exports.
Vaidio supports alerts, search, dashboards, and integrations, but it is not positioned as a complete EHS management system.
Safety teams should confirm how an identified event becomes:
Some of these functions may require an EHS platform, work-management system, VMS workflow, or custom integration.
Vaidio publishes use cases, partnerships, platform information, and selected customer comments. Detailed public industrial case studies containing named facilities, baselines, deployment periods, safety outcomes, and financial results remain limited.
Prospective buyers should request:
A controlled pilot can provide evidence that is more relevant to a specific facility than broad platform claims.
Vaidio may be relevant when an organization needs several of the following:
The platform may be particularly relevant to enterprise security teams seeking to extend an established camera and VMS environment across several security and investigation workflows.
Industrial organizations should establish whether the project is primarily owned by security, IT, operations, manufacturing, or EHS. That ownership affects analytics selection, alert response, privacy rules, implementation priorities, and success measures.
Voxel focuses on industrial safety and operational-risk use cases rather than broad physical-security functions.
Its published detections include supported camera-visible events involving:
Voxel is not positioned as a general platform for facial recognition, license-plate investigations, weapons detection, or public-space crowd monitoring.
Vaidio may therefore be relevant to a security-led analytics program, while Voxel may be relevant to an EHS or operational-risk initiative. An organization could also evaluate the platforms for separate functions within the same facility.
Voxel’s industrial risk platform organizes supported events into video evidence, trends, and reporting. Teams can convert identified risks into recommended actions, assign owners and deadlines, and review the effect of completed interventions.
Potential responses include:
These workflows address what happens after an industrial-risk signal is identified. They do not replace a VMS, general security operations center, facial-recognition system, or access-control platform.
Voxel does not use facial recognition. Its published privacy and security controls include face and body blurring, role-based access, SSO support, SOC 2 Type II controls, encryption in transit and at rest, and infrastructure hosted in ISO 27001-certified data centers.
The Carlex customer story describes a deployment developed in collaboration with United Auto Workers leadership. Management established that the system would be used for information gathering and training rather than punitive monitoring.
At its Vonore facility, Carlex reported:
Most of the reported improvement occurred within three months. The case attributes the results to Carlex’s use of Voxel data alongside coaching, signage, traffic changes, training, and management action rather than to software acting independently.
The appropriate platform depends on the primary requirement. Vaidio may fit organizations seeking broad site security, access, and investigation analytics. Voxel may fit organizations seeking camera-visible industrial-risk detection connected to corrective actions and safety reporting.
Organizations can contact Voxel directly to review supported industrial use cases, camera coverage, privacy requirements, and action workflows.
The platform and company are now marketed under the Vaidio name. IronYun Inc. USA remains the legal entity identified in the company’s current software agreement. The rebrand was announced in March 2025 alongside the ninth generation of the platform. Current product releases, including Vaidio 10.0, use the Vaidio brand.
Vaidio supports intrusion detection, loitering, perimeter monitoring, object search, crowd analysis, license-plate recognition, facial recognition, access-control analytics, and other configurable applications. It can analyze live and recorded video and help operators search across camera footage. Available functions depend on the licensed analytics, infrastructure, camera conditions, and configuration. Buyers should test their required scenarios rather than relying only on the size of the analytics library.
Vaidio publishes applications involving PPE, falls, hazardous areas, smoke, fire, leaks, and selected forklift-pedestrian risks. These functions may support industrial monitoring when properly configured and validated. Facilities should confirm alert latency, camera requirements, detection performance, and response procedures for each use case. Safety-critical controls should continue to rely on appropriate engineering systems, procedures, and human safeguards.
Vaidio is a broad AI vision platform covering security, access control, forensic search, operations, traffic, and selected safety applications. Voxel focuses on camera-visible industrial risks and workflows for reviewing events, assigning actions, supporting coaching, and reporting trends. Vaidio may be relevant to a security-led program, while Voxel may be relevant to an EHS-led risk program. Some organizations may evaluate both for separate operational purposes.
A pilot should include representative cameras, work areas, lighting, weather, object sizes, vehicle types, worker density, and routine operating conditions. Teams should document false alerts, missed events, alert latency, configuration effort, operator workload, and integration performance. Privacy settings, retention, permissions, exports, and escalation procedures should also be tested. Expansion should follow demonstrated performance for the intended use cases.
Explore AI Summary
