Frontier Firm BNY resolves client inquires 80% faster with Microsoft AI powered Eliza

BNY is a 242-year-old bank that oversees approximately $60 trillion on behalf of clients across 100 markets. The legacy bank embraced AI early on and believed firmly in its ability to reshape how financial systems operate.

BNY built an in-house AI platform, Eliza, that incorporates Microsoft platforms and tools like Microsoft Azure, Microsoft Foundry, Microsoft 365 Copilot, and GitHub Copilot.

Today, digital employees handle 10% of payment validations, client settlement inquiries are resolved 80% faster and new customers are onboarded 20% quicker with Eliza. The platform now includes over 300 solutions and 100% of employees are trained to use it.

Around the world, clients trust BNY, a financial services platform company, to custody over $60 trillion in assets. Alongside its considerable impact on the global financial system, the company has a historical perspective as one of the oldest banks in America, founded in 1784. Despite the long history of the bank, Leigh-Ann Russell, CIO and Global Head of Engineering, BNY, calls this a particularly exciting time. Why? The bank embraced AI capabilities early on. “We have always used machine learning as long as it’s been available,” she says. “Now is one of the most exciting times in my career because of how this technology has evolved and has the potential to change how financial systems work.”

“Now is one of the most exciting times in my career because of how this technology has evolved and has the potential to change how financial systems work.”

Leigh-Ann Russell, CIO and Global Head of Engineering, BNY

In-house AI platform incorporates Microsoft 365 Copilot and Microsoft Foundry

Today, AI goes by a single name at the bank: Eliza. In 2023, BNY set up an AI Hub and went to work creating an AI platform. The platform was named Eliza in honor of Eliza Hamilton, the wife of bank founder Alexander Hamilton — a philanthropist in her own right. “Our platform combines many elements that other institutions keep separate,” says Jordan Romanoff, Senior Director, Co-head of Intellectual Property, AI, and Innovation Legal and Head of Marketing and Communications Legal, BNY. Those include governance, learning resources, and AI tools and agent building capabilities. “Having Eliza as a single platform, a one-stop-shop for our employees, makes it easy to engage with Eliza and AI more broadly,” says Russell. “And, we had governance and trust baked into the platform from day one,” she adds. Employees like Romanoff use the Eliza platform as a thought partner. “I have had Eliza help me do analysis that could have taken me hours,” he says. “When you realize how it can make you smarter and enrich your analysis with additional information, it’s an a-ha moment.” Today, 100% of employees are trained on Eliza.

Eliza was built to be multi-agentic and model-agnostic, an approach that included Microsoft technology as a key aspect of the solution. “When we started to build out Eliza, we selected the Microsoft Azure cloud foundation, then added Microsoft Foundry as part of the solution,” says Sarthak Pattanaik, Chief Data and AI Officer, BNY. “The continuous execution of resiliency, security and the power of frontier models aligned with our strict needs and the strong support Microsoft has provided us, as well as the data gravity we have already with Microsoft, made it the right choice.”

“When we started to build out Eliza, we selected the Microsoft Azure cloud foundation, then added Microsoft Foundry as part of the solution. The continuous execution of resiliency, security and the power of frontier models aligned with our strict needs…”

Sarthak Pattanaik, Chief Data and AI Officer, BNY

The partnership didn’t stop there. “We have now integrated Microsoft 365 Copilot into Eliza directly,” says Russell. “With both these tools working side-by-side, our employees can be very creative with their AI uses.” Prior to the integration, BNY was an early adopter of Microsoft 365 Copilot in the financial services space, a decision that Russell describes as “hugely popular” with employees.

GitHub Copilot is another Microsoft tool that employees access via Eliza. “Our software engineers use GitHub Copilot for coding assistance,” says Saed Shonnar, Head of AI Enablement, BNY. “Close to sixty percent of all new code is being developed with AI coding assistance, leading to a ten percent increase in releases.”

“We have now integrated Microsoft 365 Copilot into Eliza directly. With both these tools working side-by-side, our employees can be very creative with their AI uses.”

Leigh-Ann Russell, CIO and Global Head of Engineering, BNY

Resolving client transaction inquiries 80% faster and onboarding new customers 20% quicker with Eliza

Eliza was always meant to be more than a point solution. Today, the platform encompasses more than 300 solutions. “The technology has very broad applications throughout the financial services value chain,” says Shonnar. “We see applications and use cases in every aspect of what we do.” One of those use cases is client onboarding, a particularly complex process in the highly regulated world of financial services. It was a process ripe for an AI-driven solution. “Twenty-five percent of all of our new onboardings were assisted with AI this year and that has resulted in a twenty percent faster onboarding process on average for these clients,” says Shonnar.

Resolving client inquiries was another process that lent itself to improvement with the Microsoft-backed Eliza platform. “Many of our transactions result in inquiries from our clients,” Shonnar explains. “More than ten percent of these inquiries were resolved or assisted by AI and that has resulted in eighty percent faster processing of these inquiries.” In the past, inquiries would have involved a human employee and potentially multiple stakeholders to resolve a single question.

BNY has applied agentic AI to another time-consuming process at the bank. Client onboarding processes are mandatory to verify user identities and review risk. “This research element of the onboarding process is a common challenge that involves document processing and decision-making that can lead to weeks of back and forth,” says Pattanaik. BNY created an agentic workflow that can review new documents, leverage previous documentation, pursue new data, and ultimately help make a risk decision. Today, this AI-enabled workflow is helping the bank onboard 25% of new customers with 20% faster completion of certain processes. “It’s an amazing use case,” says Pattanaik.

“Our software engineers use GitHub Copilot for coding assistance. Close to sixty percent of all new code is being developed with AI coding assistance, leading to a ten percent increase in releases.”

Saed Shonnar, Head of AI Enablement, BNY

Delegating time-consuming tasks to digital employees

BNY has progressed to embracing fully-fledged digital employees. Russell explains; “We have digital employees in Eliza who work semi-autonomously across very complex processes.” One of those processes is payment validation. “When a payment is going from point A to point B, sometimes there are instructions in the middle that are missing or incomplete,” says Romanoff. “We developed a solution that helps repair those instructions and allows the payment to proceed without having to go through a time-consuming review. This has sped up our ability to send payments around the world. Today, that digital employee handles over ten percent of our payment repair issues around the world.”

Already, Eliza is entrenched at the bank and AI has become deeply embedded in how people work and solve problems. “We now have one of the broadest adoptions of AI in financial services and that has been led by our employees who have come up with great ideas,” says Russell. Those great ideas are having an impact on BNY clients. Pattanaik explains; “AI helps us to provide our clients with insights that can help them be successful in their businesses,” he says, adding that the technology helps BNY create bespoke products for customers and helps keep the ecosystem safe, further building customer trust. Next, BNY will add Work IQ and Microsoft 365 Copilot Cowork to the mix. “Today we use the Microsoft Graph and we’re excited to incorporate Work IQ and Copilot Cowork so that the interface is even more seamless and we can do more,” says Shonnar. “It is critical for us to be able to tap into Microsoft systems whenever we’re using AI.” Romanoff agrees: “By leveraging our M365 connectivity, it means faster responses and deeper insights that are personalized to their actual workflows.”

“Today we use the Microsoft Graph and we’re excited to incorporate Work IQ and Copilot Cowork so that the interface is even more seamless and we can do more.”

Saed Shonnar, Head of AI Enablement, BNY