CaseTrace System
Identity-First Investigative Infrastructure
Operated by Guard Industry Services
System Definition
CaseTrace is an identity-first investigative documentation and correlation system designed for licensed security operations acting as operator-of-record.
It is built for environments where identifying recurrence, preserving structured records, and maintaining supervisory authority must occur within clearly defined legal and operational boundaries.
CaseTrace is infrastructure.
It is not a recommendation engine.
It is not an enforcement system.
It is not an automated decision-maker.
It structures investigative information and surfaces probabilistic recurrence signals while preserving human authority.
All operational decisions remain discretionary.
I. System Classification
CaseTrace is a Class A — Identity-First System.
Identity-first systems are defined by the following structural properties:
- Identity inference is central to system logic
- Correlation is a core functional layer
- Recurrence detection is intentional and explicit
- Authority resides with the licensed operator
- Escalation is policy-driven, not software-driven
Identity-first systems differ from documentation systems because identity inference is embedded in system design.
CaseTrace does not conceal this structure.
It declares it.
II. Architectural Model
CaseTrace operates across three structured layers:
- Documentation Layer
- Correlation Layer
- Supervisory Authority Layer
Each layer is intentionally separated to preserve role clarity and authority boundaries.
1. Documentation Layer
Structured Incident Preservation
Security personnel document incidents using standardized, time-stamped reporting tools.
Each incident record may include:
- Narrative observations
- Location metadata
- Time metadata
- Observed subject characteristics
- Vehicle descriptors
- License plate entries
- Uploaded images or video stills
- Internal annotations
- Guard identification and submission logs
Design characteristics:
- Records are cryptographically time-stamped
- Edits are logged and preserved
- Narrative authorship is maintained
- Storage is encrypted
- Access is role-based
The system records what was observed.
It does not interpret it.
Documentation is preserved as historical record, not as legal conclusion.
2. Correlation Layer
Probabilistic Recurrence Detection
The correlation layer evaluates documented data using structured similarity models.
Correlation types may include:
Facial Feature Similarity
- Feature vectors derived from stored media
- Similarity scoring against historical case vectors
- Threshold-based surface display
- Case reference linking
License Plate Recurrence
- Structured plate normalization
- Recurrence frequency detection
- Prior documented occurrence display
- Associated case references
Descriptive Pattern Recurrence
- Structured attribute comparison
- Narrative metadata indexing
- Cross-incident descriptive similarity signals
Cross-Sector Correlation (Optional via GDN)
- Correlation signal exchange between participating operators
- Referenced case identifiers only
- No raw narrative transfer
- No identity confirmation
Correlation results are displayed as:
- Possible correlation
- Referenced in case
- Prior documented occurrence
- External correlation available
The system does not:
- Confirm legal identity
- Declare match certainty
- Establish probable cause
- Trigger enforcement
- Mandate escalation
Correlation signals are probabilistic indicators requiring human evaluation.
Matching Threshold Model
Similarity thresholds are configurable at administrative level within predefined safe parameters.
The system may allow:
- Conservative threshold tuning (lower false positives)
- Aggressive threshold tuning (higher sensitivity)
- Supervisory review gating before recurrence visibility
- Cross-sector enable/disable control
Threshold changes are logged and auditable.
The system does not self-adjust thresholds autonomously.
III. Supervisory Authority Layer
Authority remains with licensed personnel and supervising operators.
Supervisors may:
- Review historical recurrence
- Examine side-by-side media comparisons
- Evaluate similarity score distributions
- Access structured incident history
- Escalate internally per policy
- Preserve documentation for legal process
The system does not:
- Escalate automatically
- Contact law enforcement
- Generate external alerts
- Initiate detention protocols
- Override policy
Escalation is an operational decision.
Not a software action.
IV. Edge Processing Architecture (Optional)
When deployed with edge devices:
- Facial similarity processing may occur locally
- Entry-point recurrence detection may operate in real time
- Feature vectors may synchronize securely to central storage
- Local processing reduces latency
Edge devices:
- Do not perform license plate recognition unless separately enabled
- Do not initiate enforcement
- Do not override supervisory authority
- Do not generate autonomous alerts to external entities
Edge enhances visibility.
It does not assume control.
V. Global Data Network (GDN)
GDN enables optional cross-sector correlation among participating licensed operators.
If enabled:
- Feature vectors may be evaluated across participating systems
- Recurrence signals may reference external case identifiers
- No narrative text is shared
- No raw incident records are transferred
- No legal identity confirmation is made
GDN does not:
- Confirm identity
- Share investigative files
- Mandate inter-organizational response
- Trigger enforcement
Cross-sector signals remain probabilistic.
Participation is administratively controlled.
VI. Data Governance & Security
CaseTrace implements:
- Encrypted storage at rest and in transit
- Role-based access control (RBAC)
- Multi-factor authentication (optional)
- Immutable audit logging
- Administrative action tracking
- Correlation threshold audit logs
- Secure synchronization protocols
- Controlled data export mechanisms
Data retention policies are configurable by operator within applicable legal constraints.
The system does not:
- Sell data
- Monetize biometric records
- Share data outside authorized participation
VII. Authority & Liability Model
CaseTrace is operated by Guard Industry Services as operator-of-record.
Guard Industry Services:
- Maintains system operation
- Oversees deployment
- Manages configuration
- Defines permissible usage parameters
- Ensures compliance alignment
The system does not expand legal duty beyond the operational authority of licensed personnel.
Authority resides in human operators.
Software supports investigative visibility.
It does not assume investigative authority.
VIII. False Positive Management
Identity similarity is probabilistic.
The system is designed to:
- Display similarity indicators clearly as "possible"
- Preserve score transparency
- Require supervisory confirmation
- Avoid automatic escalation
- Allow threshold tuning
- Log review actions
False positives are expected in probabilistic systems.
The system does not conceal this.
Professional review is required for interpretation.
IX. Regulatory Posture
Deployment requires compliance with:
- Jurisdictional biometric regulations
- Local surveillance laws
- Licensing requirements
- Professional standards
- Contractual operating policies
Guard Industry Services operates CaseTrace within defined regulatory frameworks.
The system does not certify compliance automatically.
Operators remain responsible for lawful usage.
X. What The System Is Not
CaseTrace is not:
- An automated policing system
- A predictive behavior engine
- A guilt-determination tool
- A law enforcement database
- A substitute for investigative procedure
- A replacement for professional judgment
It is investigative infrastructure.
XI. Structural Summary
CaseTrace:
- Structures documentation.
- Surfaces recurrence.
- Preserves record integrity.
- Maintains human authority.
- Avoids autonomous escalation.
- Operates within operator-defined boundaries.