Kriyam AI — agentic threat model

Uses specialized computer vision and facial recognition models for Video KYC. Primary threats include adversarial evasion attacks (e.g., deepfakes, presentation attacks) and document spoofing designed to bypass identity verification.

Processes highly sensitive identity documents, video streams, and digital signatures. Key threats include data exfiltration of PII, unauthorized access to stored verification records, and poisoning of verification reference databases.

Orchestrates automated field investigation workflows and real-time document verification. Vulnerabilities include insecure tool integration where document parsing or signature capture APIs could be manipulated to bypass verification steps.

Not certain from the listing — as an open-source platform, deployment security depends heavily on the self-hosting environment. Threats include container escape, insecure API endpoints, and exposed database ports hosting sensitive KYC data.

Provides compliance dashboards and analytics. Threats include blind spots in detecting spoofing attempts, lack of robust logging for failed verification attempts, and drift in facial recognition accuracy over diverse demographics.

Focuses heavily on compliance and secure consent capture. Threats include weak authorization controls allowing unauthorized access to compliance reports, and cryptographic weaknesses in the digital signature implementation.

Not certain from the listing — there is no explicit mention of multi-agent orchestration or marketplace integrations. Threats would involve unauthorized third-party agents accessing the KYC verification pipeline.