For Indonesia’s fast-evolving financial sector, this shift offers extraordinary opportunities, alongside complex challenges that demand urgent strategic focus.

Traditional AI systems require detailed instructions to complete specific tasks. In contrast, agentic AI systems can operate independently to achieve goals. These agents can interpret broad objectives, create strategic plans, and execute workflows across departments without manual inputs at every stage. Two features set them apart: “under specification”, accomplishing tasks with minimal guidance, and “long-term planning”, making decisions that shape future outcomes.

Take customer onboarding, for example. Currently, Indonesian banks juggle multiple touchpoints, documentation and compliance checks. An agentic AI system can manage the entire onboarding journey: Verifying identity, screening for risk and antimoney laundering (AML), and adapting to individual customer needs in real time. This creates a frictionless experience, reducing dropout rates and improving compliance at scale.

Indonesia’s financial landscape is uniquely positioned to benefit from agentic AI. The country’s geographic spread, multilingual population and regulatory complexity pose a level of operational challenge that traditional systems often fail to manage. But agentic AI thrives in this kind of multifaceted environment.

As IBM’s Agentic AI in Financial Services whitepaper outlines, applications span three critical areas. One area of impact is customer engagement and personalization. Agentic AI enables banks to deliver hyper-personalized services based on cultural preferences, income levels and behavioral patterns.

For the underbanked population, it can analyze alternative credit data, communicate in local languages and guide users through financial education and micro-lending options in a way that is accessible and culturally relevant.