Oracle has expanded its enterprise artificial intelligence (AI) capabilities with 25 prebuilt agentic applications designed to execute business processes across finance, human resources, supply chains and customer experience, as the company moves towards what it describes as a "system of outcomes".
Announced at Oracle AI World 2026, the applications operate within Oracle Fusion Cloud Applications and are powered by Oracle Fusion Claw, an execution framework that enables AI agents to coordinate tasks, manage workflows and complete activities under organisational policies.
The new applications are available across Oracle Fusion Cloud Enterprise Resource Planning (ERP), Human Capital Management (HCM), Supply Chain & Manufacturing (SCM) and Customer Experience (CX).
They support functions including accounting reconciliation, workforce planning, payroll management, procurement, supply allocation and customer service monitoring.
Oracle said the approach is intended to move enterprise software beyond maintaining business records towards executing processes based on defined organisational objectives.
The company also introduced Oracle Fusion Companion, which allows users to describe business goals while the system identifies and coordinates the necessary transactions, approvals, workflows and AI agents.
Employees can access these capabilities through supported AI interfaces, including ChatGPT, or through Oracle Fusion Agentic Home, a new environment for assigning tasks and monitoring their progress.
According to Oracle, execution remains governed by the permissions and controls established within Fusion Applications, even when requests originate from external AI interfaces.
The company has also introduced the Fusion Setup and Configuration Registry, a publicly available repository designed to make Fusion application configurations accessible to AI agents.
The registry enables AI assistants to translate business requirements into implementation instructions, configure applications and conduct automated testing, subject to expert review before deployment.
Oracle's AI Success Navigator, previously known as Oracle Cloud Success Navigator, is also receiving additional agent-based capabilities for application design, configuration, testing and integration.
The company cited early adopter Lexington Health, which reported workload reductions of up to 80% for certain design and configuration activities during implementation.
Alongside these developments, Oracle outlined its Deep Trust AI Architecture, which places data access controls and business rules within the database rather than relying exclusively on application-level restrictions.
The architecture includes Oracle Deep Data Security, controlled data interfaces and SQL Assertions intended to prevent unauthorised actions by users or AI agents.
Oracle is also developing a Business Semantic Layer and an AI Data Catalog for its Autonomous AI Lakehouse, allowing agents to interpret business metrics and access information across different data environments.
These lakehouse capabilities are expected to become available within the next 12 months.
While Oracle has made its 25 agentic applications available, some supporting data management features remain under development. The company has not disclosed comprehensive customer adoption figures or independently verified performance results for the expanded portfolio.
Disclaimer: This article may include information derived from interviews, press releases, public statements, research, company communications and other publicly available or third-party sources. Such material may be summarised, paraphrased or contextualised for journalistic and editorial purposes. All rights in third-party content remain with their respective owners.