Identity verification and fraud screening with deepfake detection for people and AI agents

DeepIDV consolidates the checks a regulated business runs at onboarding into one workflow rather than a chain of separate vendors. The platform ingests government-issued identity documents, a selfie or video capture, phone and carrier data, address records, and banking or credit history, then evaluates them against watchlists, sanctions and PEP databases, and background sources. What comes back is a structured decision with the risk signals behind it, delivered through a hosted flow or a JSON API, along with the evidence record compliance teams need at audit. Modules can be turned on individually, so a crypto exchange and an insurer can run different check sequences on the same infrastructure.

The element the company emphasises is presentation-attack defence. Liveness and face matching are paired with detection models aimed at deepfaked video, face swaps, and manipulated or wholly synthetic documents, which matters as generative tooling makes a convincing forged applicant cheap to produce. DeepIDV also offers verification of AI agents acting on a person's behalf and exposes an MCP server so agentic systems can call the checks directly. The company reports SOC 2, ISO 27001, GDPR, and CCPA alignment. Founded by Shawn-Marc Melo, DeepIDV closed a $1M seed round in March 2026 and opened a San Francisco office alongside its engineering team in Toronto.