TrustDecision, a Singapore-headquartered risk intelligence provider, has announced the expansion of its agentic AI capabilities for financial institutions across Southeast Asia, bringing AI-driven automation to credit risk and fraud detection workflows. The announcement was made at Money20/20 Asia in Bangkok on 24 April 2026.
The capabilities are embedded within TrustDecision's unified risk decisioning platform and include two domain-specific AI agents. The Investigation AI agent supports fraud and money laundering case analysis, covering data retrieval, fund tracing, behavioural analysis, and report generation to enable faster, more consistent investigation workflows through human-AI collaboration. The Rule Mining AI agent converts investigation outcomes and risk signals into deployable rules and model features, enabling institutions to evolve from case-by-case handling to systematic, continuously improving risk strategies. Together, the agents form a closed-loop system connecting detection, investigation, and strategy optimisation.
Governance, auditability, and human oversight
TrustDecision frames the primary challenge for financial institutions as no longer model capability but governance, control, and explainability in live operational environments. The company distinguishes between use cases where autonomous AI agents are appropriate, high-volume, lower-risk tasks such as transaction flagging and application pre-scoring, and critical decisions that require human oversight and clearly defined controls.
Southeast Asian adoption remains uneven, with some markets moving quickly in digital banking and real-time decisioning while others continue to build foundational infrastructure. Thailand is identified as a market where the shift is becoming more immediate, with the first batch of virtual banks expected by mid-2026, creating pressure to build a digital-first approach. These AI-native operating environments are simultaneously efficient and compliant.
Commenting on the news, Henry Li Nan, Managing Director for Singapore, Malaysia and Thailand at TrustDecision, noted that institutions are adopting a measured approach with human-AI collaboration at the core, with higher-risk decisions retaining human oversight and a strong emphasis on model transparency within regulatory boundaries.
Adding to this, Dr Simon Liu, Chief Data and AI Officer at TrustDecision, said the more important question in regulated financial services is how a system behaves once it is live, including where autonomous decisions are appropriate and how institutions detect when a system operates outside defined boundaries.
