TD Bank Unveils Notable Agentic AI Application for Clients

In a significant leap for banking technology, TD Bank has introduced its first agentic AI application, a development that could drastically reduce the time needed for mortgage and Home Equity Line of Credit applications. This tool promises to cut the application process from an average of 15 hours to less than three minutes, simplifying operations and improving customer experience.

Pioneering Work by Sandra Aziz and Team

Sandra Aziz, a Senior Machine Learning Engineer at TD’s Layer 6, led the project alongside a team of scientists and engineers. Aziz describes the initiative as a venture into uncharted territory, saying, “We built where nothing else existed. Everything is new.” This statement encapsulates the ambitious nature of the project, which aims to integrate advanced AI capabilities into everyday banking functions.

The agentic AI application automates the application process by scanning client documents, calculating income, and validating it against policy requirements. It performs consent checks and verifies client income, ultimately generating a memo for underwriters that summarizes its findings. Aziz noted that the technology can consolidate information from multiple data points, delivering relevant insights to human operators efficiently.

Transforming Client Experience

The implications of this technology extend beyond mere efficiency. Aziz envisions a future where banking decisions, which typically took days, could be executed in a fraction of the time, fundamentally altering the client experience at TD. The initial results of the agentic AI application have been promising, indicating that automation could redefine how clients interact with the bank.

Since the launch of this AI model in May, TD has already begun to see the benefits, with clients experiencing faster processing times for their applications. This development positions TD as a leader in AI banking solutions in Canada and highlights a broader trend in the financial industry toward embracing advanced technologies to enhance service delivery.

The Evolution of AI at TD

Aziz's journey at TD reflects the bank's evolving approach to artificial intelligence. She recalls when the team was initially focused on statistical analysis, gradually transitioning into analytics, predictive modeling, and now, agentic AI. “When I joined, we were called the statistical team, and then we became more of an analytics team doing predictive modeling, and then we started to embrace AI,” she explained. This evolution underscores TD’s commitment to innovation in financial technology, with agentic AI representing the next step in this journey.

As TD continues to refine this application and explore further advancements in AI, the bank is positioned to significantly enhance its operational efficiency and client satisfaction. The focus on agentic AI could serve as a blueprint for other institutions looking to modernize their processes and improve customer interactions.

Aziz is enthusiastic about the potential of agentic AI, stating, “The agentic AI agent will summarize different information from different data points and provide the right information to a human.” This sentiment highlights the consequential capability that such technologies hold for the banking sector, paving the way for a future where clients can access banking services more swiftly and smoothly than ever before.