Oracle Announces Fusion Agentic Applications

Oracle launched Fusion Agentic Applications, a new class of collaborative AI agents built directly into Oracle Fusion Cloud Applications to autonomously reason, decide, and execute enterprise workflows.
新製品NQ 92/100出典:PR Times

📋 Article Processing Timeline

  • 📰 Published: April 14, 2026 at 21:20
  • 🔍 Collected: April 14, 2026 at 13:01
  • 🤖 AI Analyzed: April 19, 2026 at 16:53 (123h 51m after Collected)
Oracle AI World, London - April 14, 2026

(This document is a translated summary of a press release announced by Oracle Corporation in the US on March 24, 2026.)

Oracle today announced "Fusion Agentic Applications," a new class of enterprise applications comprising a collaborative team of specialized AI agents designed to be outcome-driven, proactive, reasoning-based, and built for operation within enterprise environments. Embedded in "Oracle Fusion Cloud Applications," the "Fusion Agentic Applications" can securely access the context of integrated enterprise data, workflows, policies, approval hierarchies, permissions, and transactions to make decisions and execute them within business processes. Unlike copilots, AI assistants, or other AI add-ons, "Fusion Agentic Applications" are built natively into transaction systems, allowing them to be executed at enterprise scale, in real-time, and under full governance.

Steve Miranda, Executive Vice President of Application Development at Oracle, stated: "The current reality, where much time is spent on process management rather than creating outcomes, cannot meet the speed, complexity, and expectations required by modern business. With 'Fusion Agentic Applications,' enterprise software evolves from a mere system of record to applications that can reason, make decisions, and actually take action to achieve defined business objectives. This is a massive step forward for the industry, helping our customers achieve results faster, focus their valuable time on strategic activities, and redefine how work gets done."

"Fusion Agentic Applications" are designed to coordinate reasoning and execution, continuously driving business outcomes and expanding upon those results. They are composed of teams of AI agents with distinct roles, expertise, and decision-making authority, capable of autonomously determining why, when, and how to advance work toward specific goals. "Fusion Agentic Applications" operate completely within the existing "Oracle Fusion Cloud Applications" security framework, autonomously executing routine actions within guardrails. It then surfaces only the exceptions, trade-offs, and critical decisions where human judgment significantly impacts the outcome to the user. By uniquely blending a system of record with a system of outcomes, organizations can gain the following benefits:

Focus on Achieving Outcomes: "Fusion Agentic Applications" focus on achieving specific business objectives, coordinating reasoning and execution across the entire "Oracle Fusion Cloud Applications" suite to continuously drive work forward.

Shared Context Across Processes: "Fusion Agentic Applications" can maintain a persistent shared context across time and multiple steps. Because agents remember intent, history, past decisions, and current circumstances, it drastically reduces the effort required by users to explain or reconstruct context at each step as work progresses.

Continuous Reasoning and Adjustment: "Fusion Agentic Applications" constantly reason, understand the nuances of each situation, evaluate trade-offs, and execute actions. Furthermore, when situations change, they re-evaluate, ensuring that work is always moving toward the goal rather than ending as just a completed task.

Enterprise-Class Governance and Auditability: "Fusion Agentic Applications" safely execute AI-driven workflows through role-based access controls, approval frameworks, and mechanisms to track all operational history. Since the entire flow of what actions were taken at which stage can be grasped, it is possible to clearly explain who made what decisions.