Finance teams know what needs to happen but don't have the time to make it happen. Pleo is rolling out a set of agentic AI tools designed to close that execution gap, as its own research finds that 78% of finance professionals believe the future of their function depends on a solid understanding and use of AI, yet 68% say AI skills and training are severely lagging.
The demand for AI in finance is clear, but adoption has been slow. Many professionals are seeking out AI for Finance training to build the skills that are currently missing.
According to Pleo, 55% of UK finance leaders have not had a full uninterrupted day on strategic finance work in the past month. The daily reality for already-stretched teams is a chain of manual steps for every spend decision, creating delays and eating into time that should go toward higher-level analysis.
Traditional AI tools have not solved this because they still require finance leaders to drive the process. Agentic AI, by contrast, executes actions while the finance leader supervises, rather than micromanaging. Pleo's move into AI Agents & Automation reflects a broader industry shift toward systems that act rather than just advise.
Pleo is building a suite of agents that run in the background, handling routine tasks so finance leaders can focus on decisions that move the needle. The Policy Agent is already live, enforcing company spend rules in real time by checking every expense and purchase against policy. It will be joined by four more agents:
Pleo's five AI agents
- Expense Agent: captures card transactions, finds receipts, writes memos, applies general ledger codes and submits claims.
- AP Agent: processes invoices from email, creates records, codes them, schedules payment and tracks approval status end to end.
- Treasury Agent: optimises cash flow, schedules payments and surfaces budget risk before it becomes a problem.
- Accounting Agent: codes transactions, reconciles accounts and closes the books.
From summer 2026, Pleo's agentic AI tools will be available in two ways. The Pleo-provided agent works inside Slack, Microsoft Teams, WhatsApp and SMS, requiring no new app. When an expense needs a receipt, it looks in the employee's inbox first; when a budget threshold is crossed, it notifies the right person. For organisations already using personal AI agents, Pleo's MCP interface connects them directly to financial data, so employees using tools such as Claude, ChatGPT or Microsoft Copilot can query spend, process expenses and run reports without switching tools.
Pleo's shift to composable infrastructure is about more than saving time on copy and paste. It's about informing every insight with knowledge built for European complexity. By next year, spend decisions will be handled by coordinated agent networks, with humans overseeing outcomes while AI executes the work. Looking further ahead, Pleo aims to become the operating system for company spend, automatically controlling, paying, recording and optimising every financial decision.
"High-level financial strategy will always be something best suited to the experts," Pleo says. "While agentic models can enrich their insights and actions, the responsibility for setting direction will remain with the people using them."
Why this matters for finance teams
Agentic AI changes the role of finance professionals from execution to supervision. Instead of spending hours on manual tasks like chasing receipts or coding invoices, teams can set the rules and let agents handle the work. This does not replace finance leaders; it gives them back the time to focus on business decisions that actually require their expertise. For finance teams that have struggled to adopt AI because traditional tools added more to their plate, this shift offers a path to finally using AI to reduce the load, not increase it.
Your membership also unlocks:
