About a-team Marketing Services
The knowledge platform for the financial technology industry
The knowledge platform for the financial technology industry

A-Team Insight Briefs

Oracle Adds Agentic AI to Investigation Hub for Financial Crime

Subscribe to our newsletter

Oracle Financial Services has added a broad set of agentic AI capabilities to its Investigation Hub Cloud Service, aimed at helping financial institutions automate parts of their investigative workflow and reduce manual effort when tackling financial crime.

The new AI agents are designed to assist investigators by gathering evidence, surfacing key insights, and generating detailed case narratives—tasks that often require significant time and effort. By handling these elements automatically, the system allows investigators to focus on the more complex aspects of cases, particularly those involving sophisticated criminal schemes.

Jason Somrak, head of financial crime product strategy at Oracle Financial Services, described the development as a “paradigm shift in financial crime investigations,” noting that Oracle’s approach allows AI agents to “follow investigative plans, collect evidence, and recommend actions while providing investigators with robust narratives documenting the findings.” According to Somrak, this process “enables firms to drive consistency in decision making and thoroughly investigate all risks automatically.”

The move reflects growing demands on financial institutions to detect and respond to increasingly complex financial crime threats, all while navigating heightened regulatory expectations. Traditional investigation methods—often reliant on manual data collection and analysis—can be slow and inconsistent, exposing firms to risks from bad actors and regulatory risks of non-compliance.

Unlike AI chatbots that depend on investigators asking specific questions, Oracle’s AI agents are built to proactively analyse alert data, identify connections (such as matches with sanction lists), and generate comprehensive narratives that summarize each case. The goal is to provide investigators with clear, relevant information to support more informed and timely decisions.

These AI-driven features form part of Oracle’s broader suite of financial crime and compliance tools, which are increasingly leveraging generative AI to improve the speed, consistency, and reliability of financial investigations.

Subscribe to our newsletter

Related content

WEBINAR

Recorded Webinar: Sponsored by FundGuard: NAV Resilience Under DORA, A Year of Lessons Learned

The EU’s Digital Operational Resilience Act (DORA) came into force a year ago, and is reshaping how asset managers, asset owners and fund service providers think about operational risk. While DORA’s focus is squarely on ICT resilience and third-party dependencies, its implications extend deep into core operational processes that are critical to market integrity, investor...

BLOG

MiFIR Schema 1.4.0 Rollout: Testing Clarity Still Pending – April Deadline Remains

As of mid-February 2026, the European Securities and Markets Authority’s (ESMA) MiFIR reporting webpage continues to indicate that a dedicated test environment for updated transparency messages would open in February, with exact dates to be confirmed in January. No detailed testing calendar has been published at the time of writing. The result is a compressed...

EVENT

Data Management Summit London

Now in its 16th year, the Data Management Summit (DMS) in London brings together the European capital markets enterprise data management community, to explore how data strategy is evolving to drive business outcomes and speed to market in changing times.

GUIDE

Applications of Reference Data to the Middle Office

Increasing volumes and the complexity of reference data in the post-crisis environment have left the middle office struggling to meet the requirements of the current market order. Middle office functions must therefore be robust enough to be able to deal with the spectre of globalisation, an increase in the use of esoteric security types and...