How Google Cloud is Bringing Agentic AI Capabilities to HSBC

Among the giants banking on AI for finance is HSBC.

HSBC will deploy Google Cloud across its operations in a partnership that could show how large language models integrate into regulated industries at scale.

The multi-year agreement provides HSBC with access to Gemini models and the Gemini Enterprise Agent Platform.

Engineering teams from both Google Cloud and Google DeepMind will work directly with the bank to build AI tools across three initial deployment areas.

According to HSBC, the partnership aims to deliver more than 200 AI use cases within two years.

Gemini platform enters banking operations

The collaboration brings Google's agentic AI capabilities into HSBC's infrastructure.

"Our partnership with HSBC is a blueprint for the future of the financial services industry," says Thomas Kurian, CEO of Google Cloud

"By accelerating AI adoption built with Gemini, our Gemini Enterprise Agent Platform, forward-deployed engineers and Google DeepMind's research expertise, HSBC is building a more intelligent, resilient and responsive bank that can create meaningful value for its customers." 

HSBC currently runs more than 600 applications on Google Cloud infrastructure.

The new agreement extends that relationship into generative and agentic AI systems.

The bank has identified initiatives it expects could generate more than US$100m through revenue or efficiency improvements.

Three deployment areas identified

HSBC has selected wealth management as one priority area for AI implementation.

The bank plans to use AI-driven insights to inform relationship managers who provide advice to clients.

The system aims to deliver tailored recommendations in real time while maintaining security protocols at each stage of client interaction.

Financial crime and risk management represents the second deployment area.

HSBC will combine gen AI with agentic AI to build detection architecture designed to identify risk at earlier stages.

According to HSBC, the system will monitor the near one billion transactions the bank processes monthly for financial crime indicators. The bank expects to cut intervention times in half when risk is detected.

The third area focuses on an AI assistant for frontline staff. The tool aims to reduce time spent on administrative tasks and meeting preparation.

According to the bank, work that previously took hours can be completed in minutes. The assistant will also encode regulatory procedures into an AI structure that staff can query during decision making.

AI architecture meets compliance requirements

Georges Elhedery, Group CEO at HSBC, notes that the deployment model keeps human oversight central to operations.

"AI is becoming one of the defining technologies of our time, allowing us to create a personalised experience for each customer, delivered in real time and at scale, while keeping human judgement, decision-making and accountability at the core," he says.

"A partnership like this one with Google Cloud helps us empower our colleagues with the tools they need to be future-ready, and supports our work in building a simple, agile, faster and more personal HSBC."

As part of the partnershir, Google engineers will be placed inside HSBC's operations rather than delivering AI as a standalone software product. This approach could mean closer integration between the Gemini platform and existing banking systems.