Corporate treasury teams are increasingly treating AI and automation as strategic priorities, but a new Association for Financial Professionals survey suggests the technology push is happening alongside persistent operational and skills gaps. The 2026 AFP Treasury Benchmarking Survey found that AI and automation entered treasury’s top five priorities for the first time, while cash and liquidity forecasting remained both the function’s leading priority and its biggest challenge.
Artificial intelligence is moving into the corporate treasury agenda, but the latest industry benchmarking data suggests finance organizations are not abandoning traditional liquidity responsibilities to pursue it.
The 2026 AFP Treasury Benchmarking Survey, sponsored by PNC Bank, found that 30% of treasury practitioners identified AI and automation as one of their top priorities. At the same time, 38% cited managing AI opportunities and risks as a major challenge, while 35% identified using AI to automate manual processes as a significant difficulty. The survey collected 425 responses from treasury practitioners in May 2026, primarily from U.S.-based organizations.
The numbers point to an emerging tension in corporate finance technology: treasury organizations increasingly recognize AI as strategically important, but many are still developing the data, governance and workforce capabilities required to deploy it effectively.
And despite the attention surrounding AI, the most persistent treasury problem remains much more traditional.
Cash and liquidity forecasting was identified as the most challenging treasury activity by 49% of respondents and the top overall priority by 63%.
That combination puts forecasting technology at the center of the next phase of treasury automation.
AI enters treasury’s core technology agenda
Treasury teams have long used technology for cash management, payments, liquidity planning, risk management and bank connectivity. AI adds another layer, potentially automating parts of forecasting, reconciliation, anomaly detection, reporting and decision support.
AFP’s findings suggest that transition is already underway. AI and automation ranked alongside established priorities such as cash management, liquidity planning and treasury technology for the first time in the survey. Larger organizations are driving much of the adoption.
But implementation is proving more complicated than simply adding an AI assistant to an existing treasury management system.
AI opportunities and risks were the second-most frequently cited challenging task in the survey. That indicates that treasury departments must simultaneously evaluate where AI can create value and how its outputs should be governed.
The issue is particularly relevant because treasury decisions can directly affect liquidity, funding, payments and financial risk.
A forecasting model that produces an inaccurate result is not merely an inconvenient chatbot response. It can influence decisions about cash positioning, borrowing, investment or working capital.
Cash forecasting remains the central problem
The persistence of cash forecasting as treasury’s biggest challenge is notable given years of investment in financial technology.
AFP found that 49% of respondents identified cash and liquidity forecasting as their most challenging treasury activity. At the same time, 63% said it was their top priority.
The underlying difficulty is partly structural. Treasury forecasts depend on information generated throughout an organization, including accounts receivable, accounts payable, sales, procurement, payroll and operating expenses. Data can arrive at different times, in different formats and with varying degrees of accuracy.
AI is increasingly being positioned as a way to address that complexity.
Gartner’s March 2026 guidance on AI-enabled cash-flow forecasting describes the technology as using driver-based AI techniques to generate probabilistic and explainable forecasts for liquidity planning and scenario analysis.
That emphasis on explainability is particularly important for treasury. Finance leaders need to understand not only what an AI system predicts but also which underlying factors contributed to the forecast.
Governance is becoming an AI infrastructure issue
AFP’s survey also exposes a gap between established treasury controls and newer AI capabilities.
Traditional policy areas such as cash management and bank relationship management received high effectiveness ratings in the survey. AI and emerging-technology policies, by contrast, received a substantially lower effectiveness score of 2.9 out of 5, according to the survey findings supplied by AFP.
This creates a governance challenge for organizations adopting AI within finance.
Treasury departments need policies covering issues such as data access, model outputs, human review, auditability, third-party AI services and potentially the use of sensitive financial information.
The challenge becomes greater when AI moves from analytics into automation.
An AI system that summarizes cash positions has a different risk profile from an agent that recommends liquidity transfers, initiates payments or changes a forecast used for funding decisions.
For financial technology providers, that distinction is helping drive demand for purpose-built AI rather than unrestricted general-purpose assistants.
Treasury still depends heavily on human skills
The AFP findings also challenge the idea that automation will make traditional treasury expertise less important.
Bank relationship management was the most widely used treasury skill, cited by 86% of respondents. Communication followed at 78%, while cash forecasting was cited by 77%. Collaboration and critical or strategic thinking were each identified by 75%.
The data suggests that treasury’s evolution is not simply a technology upgrade. It is also a change in the type of expertise finance organizations need.
AI can process large volumes of financial data and identify patterns, but treasury professionals still have to interpret business conditions, communicate with banking partners and operating teams, assess risk and make decisions when conditions change.
The survey found another gap at the leadership level. Ninety percent of respondents considered vision and future planning important for treasury leaders, but only 61% rated senior treasury professionals as effective in that area—a 29-percentage-point gap.
Lean teams increase pressure for automation
Treasury’s expanding mandate is also occurring within relatively small teams.
AFP found that 46% of organizations have fewer than five treasury employees. Those teams are increasingly expected to manage fraud risks, technology transformation, AI adoption and strategic advisory responsibilities alongside core liquidity functions.
That creates a practical business case for automation.
For a treasury team with limited headcount, reducing repetitive reconciliation, data preparation and reporting work can create capacity for forecasting, scenario analysis and strategic decision-making.
But automation does not eliminate the need for controls. Instead, it changes where those controls sit.
The future treasury technology stack is likely to combine automated data collection, AI-assisted forecasting, workflow automation and human review, with governance mechanisms determining which decisions can be automated and which require approval.
The next phase of fintech is purpose-built AI
The AFP survey reflects a broader evolution in financial technology. AI is becoming embedded in specialized workflows rather than existing only as a general-purpose productivity tool.
Treasury management platforms are already adding AI capabilities for cash forecasting, liquidity analysis and financial risk management. Gartner’s current market information on treasury platforms, for example, describes AI-enabled forecasting and liquidity insights as components of modern treasury-management systems.
The competitive opportunity is therefore shifting from simply providing access to AI models toward integrating AI with financial data, controls and workflows.
For treasury departments, the question is increasingly how to use AI without weakening the governance discipline that makes the function reliable.
The AFP findings suggest that transformation will happen on two tracks: improving the technology used to manage cash and liquidity while building the organizational capabilities needed to govern and interpret increasingly automated systems.
AI may be becoming a treasury priority, but the data indicates that successful adoption will depend on solving some longstanding problems first—or solving them with AI.
Market Landscape
Treasury technology is moving toward AI-assisted forecasting, automation and real-time liquidity intelligence. Yet adoption is occurring within highly controlled financial workflows where accuracy, auditability and human oversight remain important.
AFP’s research shows the scale of the transition: AI and automation have entered the top five treasury priorities, while nearly half of treasury organizations still identify cash and liquidity forecasting as their biggest operational challenge.
The market includes established treasury-management providers such as Kyriba alongside banks, fintech platforms and specialized AI vendors. The competitive focus is increasingly shifting toward connecting AI to enterprise financial data and allowing treasury teams to automate repetitive work without sacrificing control.
Top Insights
- AI and automation entered AFP’s top five treasury priorities for the first time, cited by 30% of survey respondents.
- Cash and liquidity forecasting remains treasury’s top priority at 63% and biggest challenge at 49%.
- AI governance is lagging traditional treasury controls, with emerging-technology policy effectiveness rated 2.9 out of 5.
- Nearly half of organizations have fewer than five treasury employees, increasing pressure to use automation to expand capacity.
- Treasury continues to depend heavily on communication, collaboration and strategic thinking alongside technical financial skills.
Power Tomorrow’s Intelligence — Build It with TechEdgeAI
