Deutsche Bank has deployed an on-premises artificial intelligence platform across its investment banking and capital markets teams in Europe and North America. The rollout brings firm-specific AI into workflows that depend on sensitive financial data, institutional knowledge, and strict client-facing standards.

The platform, developed by financial technology company Farsight, helps bankers work with internal documents, proprietary information, and established team processes without moving that information outside Deutsche Bank's own technology environment. Farsight announced the deployment on 13 October 2026.

The rollout reflects a broader shift in enterprise AI adoption. Financial institutions increasingly look beyond general-purpose assistants toward systems that can operate within existing workflows, draw on internal knowledge, and support business-critical materials. For investment banking teams, these capabilities matter most for presentations, transaction documents, and other client-facing work that must stay consistent with internal standards. Deutsche Bank says its teams have different requirements for the materials they produce, which makes customization central to deploying AI across a global, regulated organization.

The platform uses a conversational interface that lets employees generate and refine materials while incorporating the bank's previous work, internal standards, and team-specific processes. It is also designed to absorb user feedback over time. Deutsche Bank has not disclosed productivity figures or independently verified performance results for the rollout. Its real significance lies in the operating model: bringing AI into established banking workflows while retaining tighter control over sensitive data and institutional knowledge.

Deutsche Bank Brings AI Into Investment Banking Workflows

Investment banking relies heavily on document-intensive work. Bankers prepare presentations, transaction materials, and company analyses for clients, often drawing on information from multiple internal sources. Producing these materials takes more than drafting text. Teams must follow established formats, apply the right terminology, incorporate relevant historical work, and ensure every document meets the standards expected by the bank and its clients. Those requirements vary across business groups and transaction types, so a process that suits one team may not fit another's structure or review procedure.

General-purpose AI assistants can help with writing and summarization, but their usefulness in specialized financial workflows depends on the context they can access and how closely their outputs reflect established practices. Farsight's platform addresses this by incorporating Deutsche Bank's prior work, proprietary data, internal standards, and team-specific processes directly into the AI experience. Employees interact with the platform conversationally to generate and refine materials, drawing on relevant context rather than starting from a generic response.

The initial deployment covers investment banking and capital markets teams in Europe and North America, giving the platform a presence across major parts of the bank's investment banking operations. The announcement does not specify the total number of employees using it.

Why On-Premises AI Matters for Banks

One of the deployment's central features is its on-premises operating model. In this context, on-premises means the platform runs within the bank's own technology environment rather than relying on a standard external, cloud-hosted AI service for the entire workflow. That matters for financial institutions because their internal information often includes confidential client materials, transaction-related documents, and other sensitive business records.

Investment banking workflows can involve information subject to contractual confidentiality obligations and restrictions on access. Banks, therefore, need to weigh not just what an AI system can do, but also where it operates, which information it can access, and how its use is controlled. By deploying Farsight within its own environment, Deutsche Bank aims to maintain tighter control over sensitive information while still letting AI use the internal data and institutional knowledge needed for the work. The bank says proprietary information and workflow activity remain within its own technology environment.

That distinction matters because enterprise AI's value often depends on access to information a public-facing assistant could never reach. Still, an on-premises deployment does not automatically eliminate every security or privacy risk. Access permissions, system configuration, monitoring, and model governance remain important considerations no matter where a platform runs.

Farsight Tailors AI to the Bank's Own Standards

A key part of the platform is its ability to incorporate the bank's existing materials and working practices. Investment banking teams often develop their own approaches to structuring presentations and communicating financial information, approaches that reflect years of experience and accumulated institutional knowledge. Farsight describes its platform as firm-specific AI built around an organization's data, precedent, templates, and standards. Within Deutsche Bank's deployment, the system uses this context to help employees generate and refine materials that align with each team's requirements.

George Tierney, Managing Director and Global Chief Operating Officer for M&A at Deutsche Bank, said the bank's teams have specific standards for client-facing materials, and those standards differ between groups. He said the platform lets teams apply AI to their own workflows and materials while operating within Deutsche Bank's technology environment, calling the combination of control and customization critical to making AI work in a global, regulated enterprise.

That distinction highlights a gap between broad AI access and practical enterprise adoption. Giving employees a model is one step. Making it useful within a firm's actual processes requires the right data, context, controls, and workflow integration.

Keeping Institutional Knowledge Within the Bank

Institutional knowledge matters enormously in investment banking. Teams build experience through previous transactions, internal research, and repeated client engagements, some of it captured in documents and templates and some embedded in how teams organize their work. AI systems become more useful when they can draw on this accumulated context, but allowing a platform to use internal knowledge also raises questions about access, confidentiality, and governance.

Farsight co-founder and CEO Samir Dutta said financial institutions have spent decades building proprietary data, workflows, and judgment that shape how their teams operate. Running the platform within a bank's own infrastructure, he argued, lets that knowledge stay inside the institution while becoming part of its AI workflows. The practical value will still depend on how carefully the system distinguishes between different users, teams, and categories of information, since greater AI access to internal knowledge must not undermine existing confidentiality requirements.

Governance and Control Remain Essential

Keeping AI within a bank's technology environment addresses one part of enterprise deployment, but it does not remove the need for governance. Financial institutions must still determine which employees can access specific information, how the platform handles confidential materials, how outputs get reviewed, and how its use fits within existing policies. Banks also need to weigh the reliability of AI-generated content, since a document can follow the correct template and still contain an inaccurate statement or an outdated figure. Review procedures remain essential when AI supports materials used in client discussions or transaction processes.

The Bottom Line

Deutsche Bank's deployment points to a wider question facing financial institutions: how to capture the benefits of AI while maintaining control over sensitive information and business-critical processes. The answer increasingly involves specialized systems that combine AI models with internal data, established workflows, and organization-specific requirements, a model relevant well beyond investment banking to any financial business shaped by regulation and accumulated expertise.

For enterprise AI providers, the opportunity is to show that firm-specific platforms can deliver measurable value while preserving the governance regulated institutions require. For Deutsche Bank, the test ahead is whether this operating model translates into real efficiency gains and consistent, well-governed output across its global investment banking business.