AI Transparency & Human Oversight Notice

This Notice explains where and how SenSec LLC (“SenSec,” “we,” “us”) uses Artificial Intelligence (AI) and Machine Learning (ML) within our platform, what that means for you, and how human oversight is ensured. It supplements our Terms of Service, Privacy Policy, Data Processing Addendum (DPA), and EU Regulatory Compliance Statement.

Company: SenSec LLC, 30 N Gould St Ste N, Sheridan, WY 82801, USA
Contacts: privacy@sensec.app · ai@sensec.app (AI-specific questions)

§1Purpose of our AI systems

Our AI components are designed to assist with security operations management, not to make binding legal decisions on their own. Typical use cases include:

• Task and patrol assignment recommendations
• Incident triage, categorization, and prioritization
• Pattern detection and anomaly alerts
• Text / audio transcription and summarization of reports
• Predictive analytics (e.g., workload forecasting)

§2Human oversight & user responsibilities

• Human-in-the-loop: We expect customers to review and validate AI outputs before taking actions that could significantly affect people (e.g., employment decisions, disciplinary actions, denial of access).
• Override & appeals: Users can override AI suggestions at any time. If an individual is affected by an AI-influenced decision, they may request human review through the customer (Controller) or SenSec (if we are Controller for that data).
• Configuration & thresholds: Admins can adjust sensitivity, thresholds, and workflow rules to match risk appetite and legal requirements.
• Documentation: We provide guidance in our product docs on appropriate use, limitations, and required checks.

§3Data sources & processing

• Inputs: Customer-provided operational data (patrol logs, incidents, user actions), device / telemetry data, and publicly available or licensed datasets for model improvement.
• Pre-processing: Data may be cleaned, normalized, and aggregated; personal identifiers may be pseudonymized / anonymized where feasible.
• Outputs: Scores, categories, suggestions, alerts, summaries. Outputs may be stored for auditability and model improvement (see Privacy Policy and DPA).

§4Model lifecycle & risk management

• Risk assessment: We evaluate models for accuracy, bias, robustness, and potential rights impacts as part of our AI risk management system.
• Testing & validation: Models are tested pre-deployment and monitored post-deployment.
• Logging & traceability: Key inputs / outputs / events are logged to enable audits and incident investigation.
• Updates: Models may be retrained or tuned; material performance or behavior changes are documented.

§5Limitations of AI outputs

AI suggestions may be inaccurate, incomplete, or context-insensitive. They:

• Do not constitute legal, HR, or safety advice.
• Should not be the sole basis for high-impact decisions.
• May reflect biases in training data if not carefully mitigated; please report suspected bias or harm to ai@sensec.app.

§6Rights of individuals (EU/UK & similar jurisdictions)

Where applicable law (e.g., GDPR Art. 22, EU AI Act) grants rights concerning automated decision-making, individuals may:

• Request meaningful information about the logic involved
• Contest an automated decision and obtain human review
• Object to certain profiling activities

Requests should be directed to the relevant Controller (our customer) or to SenSec if we act as Controller.

§7High-risk AI & EU availability

If any feature qualifies as “high-risk AI” under the EU AI Act, we may delay or restrict its availability in the EU/EEA or UK until required conformity assessments and registrations are completed. See our EU Regulatory Compliance Statement and EU AI Act Declaration of Conformity for details.

§8Reporting issues & incidents

We will investigate and, where required, report serious incidents to competent authorities.

§9Changes to this notice

We may update this Notice as our AI systems evolve or as regulations change. The latest version will be posted with the “Last updated” date. Material changes will be communicated via email or in-app.