Principle 4: Security, Privacy, Compliance, and Ethics
How is data protected & AI use audited?
Responsible AI requires strong security, privacy, and compliance with federal standards. As AI adoption expands, agencies are applying it to increasingly complex use cases, including body-worn camera summarization, real-time video surveillance, license plate reading, and facial recognition tools. Although these applications deliver significant operational value, they require rigorous safeguards because the data they contain is sensitive and confidential.
When AI tools operate outside core systems this creates challenges around data security, privacy, compliance and ethical use.
The fourth pillar of Mark43’s Responsible AI Approach is: security, privacy, compliance, and ethics are foundational.
How does Mark43 protect law enforcement data?
Mark43 builds AI capabilities on its secure, CJIS-compliant cloud infrastructure. Its AI products operate within the same Mark43 platform that is relied on to manage mission-critical public safety data. All data remains within the Mark43 platform and is governed by customer agreements. AI operates directly on data in CAD, RMS and related systems, and outputs are generated and embedded within existing workflows. Hosted on AWS GovCloud and using zero-trust design, the company ensures the platform stays secure and connected, holding FedRAMP and GovRAMP High Authorization, SOC II, ISO 27001 and CISA Secure by Design.
How does Mark43 protect customers data?
- Customer data remains customer owned.
- Customer data is governed by customer agreements.
- AI models are not trained on agency data.
- Encryption and secure access controls protect data.
- Personally identifiable information (PII) is protected through filtering and redaction mechanisms.
- And Data residency controls support US based storage requirements.
Shaping AI responsibly means embedding oversight, enforcing appropriate use through guardrails, and aligning innovation with community trust.
How is Mark43’s AI use governed and monitored?
AI capabilities operate within established platform controls, ensuring usage aligns with role-based permissions, agency policies, and compliance standards. With ReportAI, governance is built into the report creation process: AI-generated draft narratives are visible within the workflow, officers review and approve content before submission, and built-in validations help enforce completeness, policy alignment, and reporting standards. With BriefAI, governance comes through traceability and reviewability: summaries are generated from authoritative RMS and case data, include citations back to source records, and surface outstanding actions so supervisors and investigators can verify key details and monitor case progress. Together, these controls support oversight and responsible use without exposing sensitive data outside the system.
Responsible AI starts with people and is sustained through strong governance, transparency, and security. Mark43’s approach enables agencies to adopt AI confidently while maintaining control, accountability, and public trust.
Book your demo today and learn more about Mark43’s Responsible AI Approach
