In this episode of Corruption, Crime and Compliance, Michael Volkov sits down with Christian Focacci, founder of Threat Digital, for their annual check-in on AI’s evolving role in due diligence and compliance. Focacci traces how AI adoption has matured from early hype and generic chatbot rollouts to more disciplined, use-case-specific tooling, while cautioning that the underlying models still hallucinate and should never be treated as an authoritative source rather than a reviewer of externally cited, verifiable information. The conversation covers the widespread problem of “shadow AI” use inside organizations without governance, the risks of letting AI creep into discretionary decision-making without human accountability, and emerging third-party risk questions as companies must now vet how vendors themselves are using AI. Volkov and Focacci also discuss the rapid rise of open-weight Chinese models, the regulatory patchwork forming at the state level (particularly around HR uses of AI), and the risk of regulatory capture favoring large AI incumbents over smaller innovators. The episode closes on a balanced note: AI is genuinely valuable for processing large data sets, triaging sanctions alerts, and boosting productivity, but only when paired with rigorous human oversight, clear documentation, and citations traceable back to verifiable source material.