AI-Powered Real-Time Object Detection for Public Safety
What it is for
Safety & Security
- Enhances operational situational awareness by detecting objects, people, or motion-based events in real time, such as a person or a vehicle entering a restricted area. Automated notifications are sent to designated monitoring staff.
Enforcement
- Supports detection of unauthorized access to restricted areas and enables alert-based responses by Emergency Management and Technology Operations Center personnel.
- Run by
- Office of Technology and Innovation (OTI)
- Where
- No fixed location
- Kept
- Retained limited retention period (specific duration not disclosed)
- Shared with
- Accountable organization
- Your copy
- You cannot see the data it holds about you. What you can do
What it is for
Fusus CORE Elite AI continuously analyzes live camera feeds to detect objects, people, and motion events — such as someone entering a restricted area — and automatically sends alerts to city monitoring staff. It is used by New York City's Office of Technology and Innovation in support of public safety and emergency management operations. The system processes live and recorded video, and its training data is proprietary to the vendor and not publicly disclosed.
What it collects and what happens to it
Data taken in
About a place
Anonymized data
- Live or recorded video streams from cameras connected to the FUSUS video management system, capturing geographic spaces and physical areas including restricted zones.
Biometric
Identifiable data
- Video streams may capture people — their physical presence, movement, and appearance — as part of the real-time detection of persons entering restricted areas or motion-based events.
Processing
Computer Vision
- Processes live and recorded video streams for real-time object detection, object classification, and motion-based event recognition. Detects people, vehicles, and other objects against configured alert profiles.
Classification & Prediction
- Assigns classification labels to detected objects and events (e.g., person, vehicle) and evaluates them against configured detection parameters to determine whether an alert condition is met.
What it does
Sensing (Perceptive AI)
Human decides
- Continuously analyzes live or recorded video streams to detect and classify objects, people, and motion-based events in real time. Outputs structured detection alerts and labeled video metadata for review by human monitoring staff.
Deciding (Analytical AI)
Human decides
- Classifies detected objects and events against configured alert profiles. Outputs classification labels and detection alerts; human monitoring staff decide on response actions.
Outputs
A recommendation or prediction
Anonymized data
- Detection alerts sent via email to OTI's Technology Operations Center and Emergency Management personnel when configured detection parameters are met. Alerts advise monitoring staff of potential events; human staff decide on response actions.
About behaviour
Identifiable data
- Labeled video metadata produced as output, recording detected object types, event classifications, timestamps, and detection contexts (e.g., a person entering a restricted area). This output identifies behavioral events captured on camera and may be linked to identifiable individuals.
Run by
Office of Technology and Innovation (OTI)
- The NYC Office of Technology and Innovation deploys and operates Fusus CORE Elite AI in support of Public Safety and Emergency Management. Alerts are distributed to OTI's Technology Operations Center and Emergency Management personnel.
Built by
Fusus by Axon
- Fusus by Axon builds and supplies the FUSUS video management system, of which CORE Elite AI is a component. The vendor provides the system as an off-the-shelf product with professional services, and training data is proprietary to Fusus by Axon.
Kept for
Retained limited retention period (specific duration not disclosed)
- Live or recorded video streams from connected cameras are stored locally and subject to a limited retention period. The exact duration is not specified in the register entry.
- Duration: limited retention period (specific duration not disclosed)
Shared with
Available to the accountable organization
- Detection alerts and labeled video metadata are available to OTI's Technology Operations Center and Emergency Management personnel, who monitor the alert email groups.
Not available to me
- Individuals captured in video or identified in detection alerts do not have described mechanisms to access the data held about them by this system. The register entry does not describe any individual data access process.
Stored
Stored locally
- Video streams from connected cameras are stored locally. The register entry confirms local storage but does not specify the jurisdiction or facility details beyond the NYC deployment context.
- Duration: limited retention period (specific duration not disclosed)
How to read the colours
Can it identify you?
- Anonymized data
- Data about people with the link to who is broken. Stripped of identifiers, blurred, aggregated, or noised so this system can’t reasonably tie a record back to an individual.
- Pseudonymous data
- Each person’s data is tied to a token (hash, ID, template) that lets this system recognise the same person across events, but the token itself doesn’t reveal a name. Reidentification is possible with extra information.
- Identifiable data
- The data either contains a direct identifier (name, address, account name, recognisable face or voice, plate number) or carries a token this system uses to look up legal identity during processing.
Who completes the loop?
- Human decides
- This mode suggests; a person decides what to do next. The AI is always advisory — a human is in the loop on every decision. Example: a triage tool ranks cases for a clinician who chooses which to see first.
- Human executes
- This mode decides; a person carries out the result. Example: an optimizer plans the day’s trash-collection routes, and drivers run them.
- Autonomous
- This mode decides and acts on its own. No person reviews each decision or carries out the resulting action.
Definitions from the DTPR standard. Amber is about your data, violet about who decides. The fuller the shape and the deeper the colour, the more identifying the data or the less a person is involved.
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AINYC Office of Technology and Innovation, reported 2025.
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- AI registerNYC AI Use Cases Register — Fusus CORE Elite AI
- Register entryPublished by the NYC Office of Technology and Innovation. Reference c227b70e. This disclosure was drafted with AI assistance.Schema: ai@2026-05-06-beta
What you can do
Ask about this system
Questions go to the NYC Office of Technology and Innovation, not the vendor.
Your rights
- Right to Be Informed of AI UseIndividuals present in areas monitored by cameras connected to the Fusus CORE Elite AI system may not be directly informed that AI-powered object detection and alerting is in operation. The register entry does not describe any public notice mechanism. Contact the NYC Office of Technology and Innovation for information about this system's deployment.
- Right to Algorithmic TransparencyThe general logic of the system — real-time object detection and classification against configured alert profiles — is disclosed in the NYC AI Use Cases Register. However, training data is described as proprietary to the vendor and not publicly available. For further information about how this system operates, contact the NYC Office of Technology and Innovation.
Risks and safeguards
- Civil liberties harmContinuous video-based surveillance in public or semi-public spaces raises risks of chilling effects on free movement and assembly, and potential for scope creep beyond restricted-area monitoring. The system is described as 'monitored' (human-in-the-loop), with alerts reviewed by designated OTI and Emergency Management staff rather than triggering automated enforcement actions. No further details on purpose limitation, oversight mechanisms, or audit procedures are disclosed in the register entry.
- Reputational harmObject classification and detection errors could result in false alerts linking individuals to restricted-area intrusions or suspicious events, potentially stigmatizing innocent people. The 'monitored' autonomy level means human staff review alerts before action, providing a check against automated misidentification. No specific accuracy thresholds, false-positive rates, or correction procedures are disclosed in the register entry.