RBC rolls out AI model to enhance banking tasks

The Royal Bank of Canada (RBC) has launched its own proprietary AI foundation model to provide a personalised customer experience and enhance banking tasks.

ATOM (Asynchronous Temporal Model), which was developed by RBC’s research institute RBC Borealis, was trained using large-scale financial datasets including “billions” of client financial transactions, which RBC said provided it with a unique breadth of knowledge in financial services.

The model is part of RBC’s Lumina data platform which systematically collects and curates its event data along all business lines, enabling AI scientists to process up to 10 billion transactions per minute.

Several RBC products and services already use ATOM’s capabilities, including credit adjudication.

The bank says that AI makes the process more accurate, consistent, and insightful, enabling the use of large volumes of complex data to assess credit, including transaction histories and non-traditional data sources.

ATOM will play a key role in its ambition to achieve $700 million to $1 billion in enterprise value generated from AI-driven benefits by 2027.

“ATOM represents the future of banking at RBC,” said Foteini Agrafioti, senior vice president and chief science officer at RBC. “It helps to personalise products and services at an individual level and enables us to more deeply understand our clients’ individual circumstances.

“As such, we will be able to tailor our services for each client and fully utilise the breadth and insights of RBC data while maintaining privacy and security.”

At the start of the year, RBC launched an end-to-end generative AI (genAI) tool designed to optimise operations and increase security in the financial services sector.

The platform, called North for Banking, was developed in collaboration with AI firm Cohere and is designed to provide a secure AI workspace that helps companies automate routine tasks, improve productivity, and streamline operations.