AI Tool Helps Fix Faulty Trades Amid Shift to Faster Settlement Times

As the financial industry grapples with the shift to shorter settlement times, banks and broker-dealers will soon have a new artificial-intelligence tool to fix and prevent trades that go awry during the settlement process.

Broadridge Financial Solutions Inc. is rolling out OpsGPT, a chat-bot that uses generative AI and large language model technology to analyze and resolve operational issues such as failed trades, and offer insight on the cause of the issue and suggestions for how to resolve it and prevent it from happening again. The product was developed to help firms deal with added stress and costs as the market moves to a so-called T+1 requirement later this year, speeding up settlement times for securities trades to one day from two.

“The need for operational efficiencies in areas like real-time-fails resolution is amplified in a T+1 world,” Vijay Mayadas, president of capital markets at Broadridge, said in an interview. The new tool helps with that transition by allowing users to “understand the deeper reasons of why a trade failed and being preemptive about it,” he said.

The world’s biggest financial firms have been experimenting more with artificial intelligence, spurred by the promise that it will help them boost staffers’ productivity and cut costs.

The new tool, which is in a trial period with a handful of Broadridge clients, can also be used in multiple languages, including Japanese, leveraging the power of large language models, Mayadas said. New York-based Broadridge integrated the technology on its existing, post-trade system, which clears and settles roughly $10 trillion in trades a day.

OpsGPT can identify a trade and figure out how positions are linked using Broadridge data across asset classes and geographies. It then recommends solutions for failed trades to avoid repeating the same error.