Basware has announced the integration of artificial intelligence (AI) agents into its invoice lifecycle management platform. This move expands the existing capabilities of InvoiceAI, with the goal of further automating accounts payable processes.

AI Agents for Finance

The company describes these agents as a step towards "Agentic Finance," a model in which AI systems perform financial tasks under preset controls. Jason Kurtz, CEO of Basware, stated that the future of finance involves near-perfect, touchless invoice processing, aiming for complete automation, compliance, and protection.

Agent Functionalities

Basware's agents are designed to operate within existing invoicing processes. The AP Business Agent provides contextual guidance to users handling invoices, recommending next steps based on the transaction's status. The AP Data Agent allows querying data in natural language, obtaining information without using reporting tools. For example, it is possible to ask which invoices are awaiting approval in a specific jurisdiction or which suppliers have granted early payment discounts in a given period.

Impact on Operations

The introduction of AI agents aims to reduce the volume of routine queries and manual follow-ups performed by accounts payable teams. According to Kurtz, this technology can transform workers' roles, allowing teams to focus on activities with greater strategic impact.

Adoption and Governance

A survey conducted on behalf of Basware found that 61% of organizations have implemented AI agents as experiments, while a quarter admits they do not fully understand how they would apply an AI agent in practice. Governance is a crucial aspect, ensuring that human operators retain control over authorization, compliance, and the audit trail. Basware's agents' actions pass through a central policy engine that applies business rules, compliance requirements, and risk thresholds.

Basware is developing further AI agents, including a Supplier Agent to manage invoice disputes and payment queries, and an AP Pro Agent to assist staff in resolving processing issues through a generative AI interface.

Implementation Considerations

The introduction of AI agents in accounts payable can reduce manual effort and response times, but the added value depends on several factors, including the quality of the AI, the condition of existing invoice data, the translation of business and governance rules into terms understandable to the agent, and the organization's willingness to delegate financial tasks to AI.