Finance leaders in hiring mode are including “AI” in nearly every job description these days, and with good reason. This is a welcome, and necessary, development—one that should be supported by integrating the latest evolution of this technology, agentic AI, into finance operations and talent strategies.
Empowering finance: integrating agentic AI into the talent model
Agentic AI offers many value-creation opportunities across industries to improve productivity and efficiency, speed of customer service, and decision-making. It can scale task and data volumes dynamically and adapt to new tasks without human intervention.
Finance functions can leverage agentic AI to mitigate the effects of talent shortages on the one hand, and to improve data production and analysis on the other. These opportunities have profound implications on corporate finance groups, which should be prepared to add an entirely new labor category—autonomous AI agents—to a portfolio of human talent resources traditionally comprised of full-time employees, contractors, captive shared services or outsourcing firms, and external consulting partners.
Undertaking this transformation requires a firm grasp of agentic AI, knowledge of its applications in finance activities and, most importantly, a solid game plan linked to the overall talent strategy.
Agentic AI defined
Agentic AI has emerged as the next frontier of AI deployment. While generative AI can create new text and images in response to prompts or requests, more advanced multimodal gen-AI models feature advanced step-by-step reasoning and research capabilities. These advancements, combined with the progression of more conventional automation technologies, such as robotic process automation, are supporting the emergence of agentic AI.
Agentic AI describes an AI system that employs a large language model (LLM), as well as other technologies, as part of its ecosystem to enable autonomous task planning and support reasoning to achieve human-defined objectives. Operating autonomously without continuous human intervention, agentic AI gathers and evaluates data to determine the best course of action and acts to accomplish its stated objective. Unlike gen-AI, which produces content for further action, agentic AI systems leverage an LLM to enable autonomous AI agents to process and interpret large volumes of data, create a plan, and produce outputs aligned to their assigned goals for review.
Agentic AI is not a future aspiration. According to Salesforce Chairman and CEO Marc Benioff, Salesforce customers already have an estimated 200 million AI agents in trials, and his own goal is for customers to have 1 billion agents in use by the end of 2026. In addition, according to research conducted by UiPath, 93% of US IT executives are “extremely or very interested” in applying agentic AI to their business, 45% of these executives are ready to invest in agentic AI this year, and more than 30% indicate they are planning on doing so in the next six months.
Autonomous finance scenarios
While overall AI use remains in the early stages across many finance organizations, some finance leaders are deploying gen-AI to complete standard compliance forms and bolster fraud detection and protection. A smaller, but growing, group of finance teams derives greater value from gen-AI solutions. These uses include forecasting adjustments throughout the order-to-cash cycle to deliver better cash flow management, as well as enhancements to third-party spending, sourcing management, IT rationalization and other forms of cost optimization.
Agentic AI holds the potential to generate even greater value across several finance areas, including:
- Core finance processes—Tasks and processes throughout the record-to-report (R2R), order-to-cash (O2C) and procure-to-pay (P2P) cycles are prime targets for agentic AI deployment, as are the financial close and consolidation processes. Finance teams can use AI agents to conduct three-way matches, investigate anomalies, enter data, manage unexpected transaction processing spikes, and more.
- Financial planning and analysis (FP&A)—AI agents can be used to conduct forecasting and modeling of supply chain operations, sales cycles and cash flow, providing insightful decision support. Agents can also produce analyses of the root causes of anomalies and exceptions that occurred in the previous quarter.
- Governance, risk and compliance functions—AI agents can work around the clock to review expense management activities, perform continuous auditing tasks, and conduct compliance-focused contract and policy reviews.
Cost optimization is another area in which agentic AI can deliver significant value. Specifically, early assessments suggest that the cost savings generated by onshoring agentic AI use in high-volume, highly repetitive back-office processes far exceed the cost savings generated by outsourcing the same processes to an offshore provider.
A game plan for governance, management and innovation
While it’s clear that agentic AI will transform finance work processes, AI governance must be in place and updated as needed. Working with the C-suite and other organizational leaders, CFOs should ensure that AI governance structures address AI agent training processes, data protection and misuse controls, bias prevention measures, accountability for the performance of AI agents, intervention protocols, success measures, human-in-the-loop (HIL) considerations, and other ethical guardrails.
In addition to addressing governance, finance leaders can launch their agentic AI journey by taking the following actions.
Make performance management adjustments
As finance groups begin to deploy agentic AI tools, new management techniques are needed in addition to AI governance and oversight mechanisms. It helps to view AI agents as “digital employees”—an extension of the workforce—to be supervised in a manner similar to human teams. Among key steps to take:
- Adopt a customer focus in defining the job and performance expectations of the AI agent.
- Set policies articulating the core values and guardrails for the agent’s behavior.
- Deploy metrics and measures to facilitate monitoring of the AI agent’s performance against expectations.
- Take remedial action when necessary and ensure the AI agent can learn and improve continuously.
The above steps largely mirror how organizations manage their people. While AI agents are autonomous and function without prescriptive and continuous human intervention, there must be human oversight of their performance and remedial intervention when necessary—just as with the work of humans.
Anticipate and address board concerns
As boards engage with executive leaders regarding the use and deployment of agentic AI, CFOs need to be ready to contribute to discussions regarding questions such as:
- What is our long-term vision for deploying agentic AI in our organization?
- How does its deployment fit within our overall AI strategy?
- What specific business problems and opportunities are we addressing with agentic AI?
- Do we have the necessary talent and expertise to design, develop and manage agentic AI systems?
- How does its deployment fit within our overall tech stack?
- What governance structures are in place to oversee the responsible deployment and use of agentic AI?
- How are we communicating our use of agentic AI to customers, employers, regulators and other stakeholders?
Analyze enterprisewide business cases and add value to their implementation
Beyond finance, CFOs should have a seat at the table to evaluate the ROI of agentic AI use cases throughout the organization. While many of these business cases will center on cost optimization efforts, others will offer additional value generation opportunities. As AI agents take over larger swaths of P2P and O2C work, accounts receivable and accounts payable teams will have more bandwidth (and better data) to address other strategic questions: What drives positive, long-term customer relationships? What are the key characteristics of our most valuable customers? How about our best suppliers? Answers to these questions can be shared with sales and procurement teams to drive revenue and profitability growth. As outsourcing contracts come up for renewal, CFOs also will want to ask those providers how they plan to leverage agentic AI to boost efficiency and lower costs.
A final point about those AI-heavy job descriptions: We’re seeing more finance leaders willing to accept candidates with less accounting experience if they possess more AI and advanced technology skills. This shift is notable amid a long-term shortage of accountants —and it shows that AI agents will soon become a crucial source of “alternative labor,” providing opportunities for employees with AI skills to become accountant/data scientists and improving the way the finance function operates.
