By using an AI-powered bot to collect, combine, and summarize information, fraud investigators can more efficiently weed out false flags and compile reports for actual criminal issues.

Startup SymphonyAI just launched a new tool to aid financial firms’ criminal investigations. 

Big FinServs have started to inject artificial intelligence into outdated know-your-customer and anti-money laundering processes to cut down on manual work and catch more bad actors, and Symphony’s new tool takes those efforts one step further.  

The firm’s new “Sensa Copilot” allows investigators to ask questions when they get a new alert, tap all available context and data sources, receive risk summaries, and draft suspicious activity reports, if necessary. Initial users report that the tool reduces the time to resolve a financial crime investigation for new alerts by up to 70%.

The platform relies on Microsoft Azure’s OpenAI service combined with Symphony’s own large language models to analyze customer data sets and linked external sources. The firm is investing “aggressively and boldly in unlocking the power of generative AI when it’s coupled with rich, specialized domain models and data sources,” exec Mike Foster said.  

While the Copilot feature is still in beta testing, SymphonyAI counts Mizuho’s investing banking arm and Citi among the customers for its overall Sensa platform.  

For other examples of how the likes of Morgan Stanley and Goldman Sachs are taking advantage of the buzziest technology on the block, check out Insights Distilled’s past coverage on how generative AI is shaping financial services.