Basware Launches Agentic AI to Make Accounts Payable Truly Autonomous
Basware has introduced new AI agents inside its Invoice Lifecycle Management (ILM) platform to push finance closer to touchless invoice processing. The goal is simple: faster approvals, fewer manual tasks, and decisions that stand up to audit and policy checks.
"The immediate future of finance involves near-perfect, touchless invoice processing," said Jason Kurtz, CEO at Basware. "The future involves Agentic Finance, where AI entities transact on behalf of the enterprise to drive faster, smarter decisions and real business outcomes."
From Automation to Autonomy
Many finance teams have tested AI without a clear path to ROI. In a global survey run by FT Longitude for Basware, 61% of finance leaders said their organizations rolled out AI agents largely as an experiment, and one in four still don't fully know what an AI agent looks like in practice. The message from the C-suite now: pilots are over-show impact.
Basware's platform is built to move beyond basic automation into AI-driven execution. It combines generative AI, natural language processing, and deep learning to handle the invoice lifecycle-from ingestion to reconciliation-with speed and control.
What the New Agents Do Today
- AP Business Agent - Provides contextual, real-time guidance on the next best action for each invoice to reduce friction and keep work moving.
- AP Data Agent - Answers natural language questions like "Show me all invoices awaiting approval in Germany," or "Which suppliers gave us early payment discounts this month?" to drive immediate, informed action.
These agents don't replace people. They operate like digital teammates: handling the repetitive queries so AP pros can focus on exceptions, cash optimization, and supplier strategy.
Trust by Design: Control, Auditability, and Clear Guardrails
Finance will only delegate to AI if the controls are airtight. Basware routes every agent action through a single, governed execution path within the ILM platform. A central policy engine and a set of autonomy gates enforce each customer's business rules, compliance requirements, and risk thresholds-before any action is taken.
The result: every AI decision is explainable and auditable with a complete trail. Finance teams keep full visibility and control while allowing AI to take on more of the work. As Kurtz put it, "Autonomy without trust is just risk."
Early Results: Billerud's ROI
Billerud, a paper and packaging materials company, adopted Basware's InvoiceAI early and reports fast gains. "The quality of invoices has improved considerably, and the AI continues to evolve and improve with each passing day," said Jesper Persson, Business Developer at Billerud. The efficiency wins translated into tangible cost savings and a rapid payback within months.
What's Coming Next
Basware plans additional agents to go beyond surfacing insights and move into automated resolution and cross-system actioning throughout 2026, including:
- Supplier Agent - Will handle invoice disputes and payment queries end to end: call the supplier, explain the issue, summarize next steps, and speed resolution.
- AP Pro Agent - Will use generative AI to help AP clerks resolve processing questions in real time through simple, natural-language prompts.
Why This Matters for Finance Leaders
AI agents aren't an experiment anymore. They're a practical lever for cash flow, cycle time, and compliance. Here's how to get value quickly without losing control:
- Start with measurable KPIs - Touchless rate, exception rate, cycle time, early-payment discount capture, and cost per invoice.
- Define autonomy boundaries - Set policy thresholds for what the agent can approve, route, or escalate. Use autonomy gates to enforce them.
- Tighten data quality - Clean supplier master data, PO policies, and chart of accounts. Good inputs make better agent decisions.
- Pilot on high-volume, low-risk work - Non-PO invoices under a limit, standard suppliers, or stable GL mappings.
- Keep humans in the loop - Require human review for edge cases and approvals over defined limits.
- Audit everything - Ensure every AI action has an explanation and a trail that satisfies internal audit and regulators.
- Integrate with ERP and procurement - Close the loop from invoice capture to payment with clean handoffs and shared policies.
The Business Case
Expect fewer touches per invoice, faster cycle times, and better discount capture. On the risk side, centralized policy controls and explainability reduce compliance exposure versus scattered bots or scripts. Most teams see quick wins within weeks once data policies and autonomy limits are in place.
Bottom Line
Basware's Agentic AI approach gives finance a controlled path from automation to autonomy. You get speed and savings without sacrificing compliance. And as new agents roll out through 2026, more of AP's repetitive work will shift to AI-while your team focuses on cash, policy, and supplier relationships.
Next steps: If you're planning your roadmap, these resources can help:
- AI for Finance - Practical applications, controls, and use cases across accounting and compliance.
- AI Learning Path for CFOs - Strategy, governance, and ROI frameworks for AI adoption.
Your membership also unlocks:
