Payables Agent for Invoice Ingestion, Compliance, and Control

Overview

The Payables Agent delivers automation and best practices across the invoice lifecycle, enabling near-touchless processing. By reducing manual effort, organizations lower errors, improve efficiency, and allow teams to focus on strategic activities. Over time, Payables shifts from a cost center to a value-generating function through faster cycle times, early payment discounts, and reduced leakage via better exception management and anomaly detection.

The touchless experience is built on unified invoice ingestion, automated processing, intuitive exception handling, and straight-through processing from capture to payment readiness. It standardizes and accelerates invoice handling with automated data enrichment, strong compliance controls, and full visibility. By replacing manual, fragmented, and reactive processes with proactive automation and supplier collaboration, organizations achieve faster processing, stronger compliance, improved supplier satisfaction, and better financial outcomes.

Powered by Gen AI, the Payables Agent automates invoice ingestion, compliance, and control across multiple sources and formats. It supports multilingual recognition and integrates with global e-invoicing providers. A unified configuration framework enables attribute defaulting, anomaly detection, and conversion of natural language Standard Operating Procedures (SOPs) into system rules. Enhanced image recognition and training capabilities streamline multi-format, multi-page invoice capture, while a consolidated Invoice List view simplifies exception management and improves operational efficiency.

Document IO

Document IO is an AI-powered document ingestion agent designed to improve the accuracy and efficiency of supplier invoice processing. It supports a wide range of file types and languages and leverages Generative AI and large language models (LLMs) to interpret invoice documents, extract and label relevant information, and intelligently map the extracted data to supplier invoice attributes. This enables more reliable and scalable processing of image-based invoices while reducing the need for manual intervention. As part of this release, Doc IO replaces the existing Intelligent Document Recognition (IDR) backend engine for supplier invoice attribute recognition. This change represents a technology upgrade only, aimed at delivering better recognition accuracy and supporting a more touchless invoice processing experience. There is no impact to the existing invoice submission process, email endpoints, or processing configurations, and all previously performed adaptive training remains unaffected. Suppliers can continue to submit invoices in the same way, and users do not need to change their day-to-day invoice processing flows.

The 26A release accelerates Touchless Invoice Processing through major infrastructure enhancements:

GenAI driven processing of image documents, replacing traditional approaches
- Extracts and labels data from submitted documents
- Intelligently map extracted data to invoice fields

GenAI Driven Image Documents Processing

In image invoice processing through Intelligent Document Recognition, relied on field level truth data and used Supplier as the anchoring factor. This made accurate supplier recognition critical and often limited scalability.

Interpret invoice documents, extract and label relevant information, and intelligently map the extracted data to supplier invoice attributes.
Stronger contextual understanding of documents.
Document IO replaces the existing Intelligent Document Recognition (IDR) backend engine for supplier invoice attribute recognition.
This change represents a technology upgrade only, aimed at delivering better recognition accuracy and supporting a more touch less invoice processing experience.

Business Benefit:

Improves the accuracy and efficiency of supplier invoice processing.
Supports a wide range of file types and languages.
Leverages generative AI and large language models (LLMs) to interpret invoice documents; extract and label relevant information; and intelligently map the extracted data to supplier invoice attributes.
Enables more reliable and scalable processing of image-based invoices while reducing the need for manual intervention.

Steps to enable and configure

For 26A, Oracle Development will reach out to select customers with steps to enable as per Early Adoption program. The feature is not available to be enabled outside of the early adoption program.

Tips and considerations

Identify high-volume suppliers that submit invoices in image-based formats, and onboard those related supplier invoice formats.
Set up Supplier Names and Addresses in Oracle Cloud Procurement or Oracle Cloud ERP. Make sure that they match exactly to the Supplier Addresses on the invoices.
Decide your PO format.
Verify that the PO formats on invoices match exactly to the PO formats set up in Procurement.
Analyze if there are any specific types of Suppliers that should be excluded from recognition. For example, you should exclude invoices from internal suppliers.

What is to Come

As the Touchless Payables experience continues to expand, the Payables Agent will grow along with it, with new features, functionalities, and experiences to support the evolving needs of customers. Customers can expect expansions and updates over the course of future releases. This includes creation of streams, a new Insights view providing rich and contextual summaries containing details and enabling action on operational and control issues, configuration options across both Capture and Compliance and Control, and further expansion of real-time, natural language analysis with the use of Ask Oracle.

Component	Experience
Configuration	Capture: Define configurations for Document Receiving Endpoints and Stream configuration definitions. Configure endpoints, email aliases, and other key details to support Invoice ingestion. Compliance and Control: Set up Anomalies, and Defaulting. Defaulting includes accounting, extensible attributes, tax, and other attributes to support effective compliance and control.
Streams	Monitor real-time visibility into invoice ingestion of streams with up-to-date processing statuses to review overall health of touchless operations. Review document capture review to view individual invoice detailed processing steps and results to accelerate understanding and actions taken on invoice ingestion and control.
Training	Access from the Streams page to upload new or review existing Invoice layouts for training and review the training status. Import or export training data from one instance to another
Invoice List	View all invoice management tasks, including both exception handling and querying on one place. View all invoices and statuses in a central hub, verify exceptions, and take the appropriate action on them.