As Artificial Intelligence becomes an expected capability across finance, treasury leaders in the U.S. are at an important decision point—how to move thoughtfully from experimentation to sustained, measurable value. The most successful teams are starting by stepping back and asking the right questions: which problems truly matter and where can AI make a meaningful difference?

Rather than spreading efforts too broadly, we encourage a disciplined approach to use case selection, focusing on a small set of high‑impact workflows, such as forecasting, exception management, reconciliation and fraud monitoring—areas where outcomes like risk reduction, faster decisions and capacity creation are clear.

Equally important is ensuring the foundation is in place to support responsible, scalable adoption. Data readiness remains a critical prerequisite—understanding what data exists, improving quality, normalizing sources and clarifying ownership—because AI outcomes are only as strong as the data that supports them. Strong governance allows treasury teams to move forward with confidence, helping ensure alignment with enterprise risk, privacy, control and third‑party standards as adoption scales.

Finally, early engagement with stakeholders across risk, legal, IT, procurement, compliance, HR and key external partners turns intent into execution, equipping teams with the capabilities needed to embed AI into day‑to‑day operations.

We hope these insights help our clients build strong AI foundations and navigate 2026 with confidence.

Katie Oresar, U.S. Head, Treasury and Payment Solutions Sales

AI represents a significant opportunity for corporate treasury departments to automate routine tasks, enhance fraud detection, improve operational efficiency and cut operational costs. According to Gartner, in 2026 nearly 60% of CFOs plan to increase AI investments in finance functions by at least 10%.

Most of the current use cases for AI involve experimentation for improving internal productivity, such as using virtual assistants to respond to client queries or to provide ideas on investment strategy. While many people have become familiar to some degree with such tools, implementation in an organizational context is a complex undertaking and requires building a strong foundation. Ultimately, success moves beyond experimentation to include broad adoption, measurable positive impact to the business and AI becoming part of the corporate culture.

For treasury teams, effective adoption and implementation are built upon four key elements:

1. Business Alignment

2. Data Readiness

3. Enterprise AI Governance

4. Stakeholder Engagement

A critical step in building an AI foundation is disciplined use-case selection. Companies too often focus on adopting the latest technology capabilities rather than identifying business problems that need to be solved. Treasury teams are well positioned to benefit from AI, especially given historical underinvestment in the function, but success depends on being strategic about where to focus limited time, data, and resources.

Treasury teams should be encouraged to take a fresh look at day-to-day processes and reimagine how workflows could be improved through AI-enabled capabilities. This requires not only clearly defining the problems to be solved but also articulating the business impact of fixing them – whether through

• 

  reduced risk

• 

  improved cash flow availability

• 

  faster decision making

• 

  or meaningful time and resource savings.

By grounding AI initiatives in well-defined problems with clear value at stake, treasury teams can better prioritize opportunities and focus on applications that deliver tangible benefits in their day-to-day operations.

Ensuring quality data is also crucial to effective AI adoption. That includes understanding what data you can collect and then structuring it so it can be used for practical applications that align with your business goals. The concept of “garbage in, garbage out” applies more strongly than ever—AI is not a cure for bad data.

Data is a particular sticking point for treasury teams. Many do not have their own platforms and tools that allow them to create their own data warehouse and have relied on spreadsheets that collate information from various sources, such as bank platforms and ERPs, to be able to create cash reports and analysis.

While tools such as Treasury Management Systems, or solutions like multi-bank reporting, have created data hubs for treasury, it can sometimes be incomplete both in breadth and depth, or difficult to normalize across systems. Recent industry moves towards the use of ISO 20022 standards has also helped in making data more accessible.

Treasury teams looking to adopt AI should take a data‑first approach, including:

• 

  Creating a data inventory with a clearly defined data taxonomy

• 

  Assessing and cleansing data

• 

  Normalizing sources into a common structure

• 

  Aligning enterprise‑wide data with the AI strategy

• 

  Understanding data security implications.

By building a good data foundation, AI outputs go from being merely informed assumptions to decision-grade insights.

Every company needs an AI governance framework and treasury departments should have a seat at the table to help shape it. AI governance establishes a broad set of policies around proper use of the technology. It also defines a set of processes and standards to help ensure AI systems and tools are integrated and supported as part of the enterprise infrastructure of the company. Most importantly, it helps guarantee that these solutions are safe and aligned with the company’s broader privacy, data and third-party policies. That includes addressing risks such as data bias, privacy infringement, use of third-party vendors and legal risks.

This is particularly important for treasury, where AI use cases directly impact liquidity and operations, including funding decisions, cash movements and risk-related transactions. A strong governance framework helps prevent decisions based on incomplete or inaccurate data and ensures appropriate human oversight—reducing the risk of underfunded accounts, failed payments or unintended transactions.

Key components of governance to consider:

• 

  Validating data sources

• 

  Aligning AI outputs with ethical, safety and regulatory frameworks

• 

  Assessing the integration of AI within third-party systems to support accountability and help maintain trust

• 

  Training employees on AI-use guidelines and protocols

Ultimately, governance allows companies to balance speed and safety. Because AI is constantly evolving, strong governance helps maintain these standards over time.

Stakeholder Engagement

Engaging key stakeholders early helps navigate the critical considerations required for a successful deployment of AI solutions. Cross-functional collaboration is essential and senior leadership must play an active role. Leaders set the tone and expectations for AI initiatives, hold teams accountable for measurable outcomes and ensure a coordinated rollout supported by clear communication

Involving all your key stakeholders will help you navigate the necessary considerations for a successful implementation. That includes making sure AI is being used to solve the right use cases. It also means understanding how AI can impact your operations and the benefits and risks it brings.

The key stakeholders that treasury should involve in AI efforts should include:

Risk management



Procurement



Legal



IT



Risk and compliance



Human resources



Third-party technology providers



Banks

Given treasury’s operational partnership with third-party partners such as banks and technology providers, aligning with these stakeholders and understanding their AI and technology roadmap could mean the difference for treasury in their ability to leverage AI sooner than later.

How Banking is Adopting AI

At BMO, our own AI journey spans across business functions to find solutions that both improve operational efficiency and deepen customer relationships. Our AI efforts have material impacts across risk, revenue, resilience and relationships, including:

• 

  Performing near real-time fraud detection

• 

  Enabling dynamic credit decisioning

• 

  Improving reconciliation procedures with an eye toward incorporating agentic AI to expand automation

We also recently established the BMO Institute for Applied Artificial Intelligence & Quantum to affirm our commitment to the responsible application and governance of AI, as well as the advancement of our quantum computing strategy.

AI promises to bring game-changing improvements to treasury functions across the board. The key is to make sure you have all the building blocks in place for a successful deployment and incorporation of AI into your operations.