Gemini Enterprise for Financial Services enters preview for capital markets and corporate banking, with Deutsche Bank an early user
Google Cloud has launched Gemini Enterprise for Financial Services, a packaged set of AI agents built for capital markets and corporate banking teams. Google Cloud says the product is in preview, with Deutsche Bank and CME Group among the first institutions using it.
The release centres on a Google-managed Financial Research agent which runs multi-step research tasks and cites its sources. Google Cloud said the agent ships with more than 50 skills, 13 data connectors and confidence scores attached to each output. The connectors pull licensed data from LSEG, Moody’s, S&P Global, MSCI, PitchBook, FactSet and SEC Edgar, among others.
The platform also plugs in third-party agents from Dun & Bradstreet, S&P Global and FlowX, among others, and Google Cloud named systems integrators including Accenture, Deloitte, KPMG, Capgemini and Infosys to handle deployment.
Deutsche Bank worked as a design partner on the agent, and will run the service first in its Corporate Bank division to research client needs and track market developments. It said it is also testing it for financial crime risk and scenario analysis.
“Starting in the Corporate Bank, we see significant potential to reduce manual research effort and give our teams more time for client conversations,” said Marie-Jeanne Deverdun, Deutsche Bank’s Chief Technology, Data and Innovation Officer.
Google Cloud released the financial services package alongside a separate one for legal teams. Both run on the Gemini Enterprise platform and are the first in a planned series of industry products.
Google Cloud: The numbers behind the platform
Gemini Enterprise reached nearly 90% of the Fortune 100 by July 2026, according to Alphabet’s second-quarter results. Google Cloud revenue rose 82% year on year to $24.8bn (£18.2bn) in the quarter, its fastest growth on record, driven by AI infrastructure and enterprise AI sales. The Gemini Enterprise assistant starts at $21 (£15.40) per person per month.
Named financial users of the wider platform include BNY, Citi Wealth, Lloyds Banking Group, Macquarie Bank and Signal Iduna. HSBC signed a separate multi-year AI deal with Google Cloud in June to build 200 use cases across wealth management and financial crime.
A contested market
Google Cloud enters a field already worked by rival model providers. Anthropic released ten financial services agents in May covering underwriting, KYC checks and pitchbook preparation, alongside a Microsoft 365 integration and a Moody’s data partnership. OpenAI announced a financial services tie-up with PwC the same week, while Microsoft has published its own plan for agentic banking.
Broadly speaking, these major technology providers are offering pre-built agents that handle research and compliance groundwork, with a person on hand to sign off the finished file – the human in the loop. Google Cloud said its version cuts bond portfolio risk analysis to under five minutes and pitch preparation from days to minutes.
Google’s move into financial services agents follows its work on agentic commerce, where it launched the Agent Payments Protocol with partners including Mastercard and PayPal.
Rules still unsettled
Supervisors have yet to fix the rules for autonomous systems in finance. The Financial Conduct Authority (FCA) said in March it would consider whether payments regulation needs rewriting for agentic AI, citing open questions over consent, liability and accountability.
An IMF paper in April argued that core payment infrastructure should stay rule-based as agents take on more work, warning that agents running similar models could act together and strain settlement systems. Identity checks face the same problem, with the industry debating how to tell an agent-initiated payment from a human one.
Google Cloud built audit logging and governance into the platform and grounds the agent’s outputs in cited sources, which it said meets the data provenance regulated firms require.
