Oracle logo on dark (Wikimedia Commons, public domain)
The enterprise software giant is bringing its agent-building platform to industry applications, extending a toolkit that began inside Fusion Cloud
Oracle has launched AI Agent Studio for industry applications, taking its agent-building toolkit beyond the back-office software where it started.
What Oracle is rolling out #
The announcement brings AI Agent Studio to Oracle’s industry applications. Until now, the platform’s home has been Oracle Fusion Cloud Applications, the company’s suite for running core business operations.
AI Agent Studio is where businesses design, test and manage those workers. Customers and partners can build custom agents through no-code, low-code or pro-code approaches.
The platform also supports multi-agent teams. These are groups of agents that split up a larger job across departments.
How the studio got here #
Oracle first unveiled AI Agent Studio on March 20, 2025, at Oracle CloudWorld London. It debuted as part of Fusion Cloud Applications, layered on top of more than 50 pre-built Oracle AI agents already embedded in that suite.
In October 2025, Oracle added an AI Agent Marketplace. It lets partners offer agents that Oracle has validated, turning the studio into something closer to an app store for enterprise automation.
Around the marketplace launch, the platform supported large language models from OpenAI, Anthropic, Google, Meta and xAI.
AI, tech, and the markets they move—in one daily briefing.
Daily. Free. Join 34,000+ readers across crypto, finance, and policy.
A July 2026 release introduced an AI-native builder that combined no-code, low-code and pro-code tools into one experience. Oracle calls the output “Fusion Agentic Applications.”
By mid-2026, more than 80,000 experts had been certified on the platform.
An October 2026 update added a modernized interface, visual builders and debugging tools, along with expanded connectivity features.
Why enterprises keep circling AI agents #
Oracle has framed the studio as a way to close the gap in enterprise AI adoption. Its approach is to embed agents directly inside trusted company data and established processes.
The platform has targeted functions such as finance, HR and supply chain. To address risk in these areas, agents built in the studio respect existing user access controls and keep an audit trail. If an employee cannot see certain data, the agent working on their behalf cannot either.
Oracle has also kept the studio available at no additional cost to eligible Fusion customers.
What this means for Oracle and its rivals #
Extending AI Agent Studio to industry applications widens the territory where Oracle can make its agent platform the default. Every new set of applications under the studio’s umbrella is another place a customer can build agents without leaving Oracle’s ecosystem.
By supporting models from OpenAI, Anthropic, Google, Meta and xAI, Oracle avoids betting on a single AI winner, positioning itself as the neutral layer where business data, permissions and agents meet.
The marketplace adds a network effect. More validated partner agents could make the platform more useful to customers, which in turn could draw more partners. The 80,000-plus certified experts suggest Oracle has been building the workforce to feed that cycle.
Disclosure: This article was edited by Diego Almada Lopez. For more information on how we create and review content, see our