March 24, 2026
Oracle Launches AI Agent Teams to Target Continuous Revenue Growth
Oracle has introduced “outcome-driven” teams of specialised AI agents called Fusion Agentic Applications, designed to continuously drive revenue growth across CX and other industry workflows. Built into the Oracle Fusion Cloud Applications suite, which includes its Customer Experience platform, the new applications were unveiled at Oracle AI World in London, alongside an expanded AI Agent Studio featuring a new Agentic Applications Builder. The launch marks a clear acceleration of the agentic AI push signalled at the company’s Q2 2026 earnings call.
Coordinated Teams, Not Single Agents
Unlike copilots or standalone AI assistants, Fusion Agentic Applications are powered by groups of agents each assigned distinct roles, expertise, and decision authority. Running natively within the Oracle Fusion Applications security framework, they draw on shared enterprise data. This includes approval hierarchies, policies, and transactional context to progress work autonomously, escalating only those decisions where human judgement is likely to change the outcome.
Crucially, they do not simply execute a task and stop. Oracle says the agents reason continuously, evaluate tradeoffs, take action, and then re-evaluate as conditions shift, keeping processes moving toward the objective rather than waiting for manual intervention at each stage. End-to-end traceability and role-based access controls are also built in throughout.
Steve Miranda, Executive Vice President of Applications Development, Oracle, summarises some of the benefits for its customers: “We are moving enterprise software beyond passive systems of record and providing our customers with applications that can reason, decide, and act in pursuit of defined business objectives. This is a huge step forward for the industry and will help our customers achieve faster outcomes, focus their valuable time on strategic activities, and redefine how work works.”
The 22 applications span finance, HR, supply chain, and customer experience. For CX and sales teams specifically, the Cross-Sell Program Workspace Agentic Application is designed to identify growth opportunities and drive what Oracle describes as “always-on revenue expansion”, replacing reactive campaign activity with a continuously active model. Agents maintain persistent context across multi-step processes, meaning prior decisions and intent are retained rather than requiring users to restart the thread at each stage.
Expanding the Agent Studio
Alongside the applications announcement, Oracle has updated Oracle AI Agent Studio for Fusion Applications with a new Agentic Applications Builder. The tool enables organisations to assemble outcome-focused agentic applications from Oracle, partner, and external agents using natural language, without requiring traditional coding skills. Other additions include improved workflow orchestration for multi-step agent processes, content intelligence that brings unstructured data into agent workflows, contextual memory so agents retain understanding across interactions, and an ROI dashboard tracking time saved and cost reductions per deployment. Oracle reports more than 63,000 certified experts are now trained in AI Agent Studio.
A Market Moving in Step
Oracle’s announcement reflects a broader industry shift from rule-based automation to genuinely agentic AI, with Gartner projecting that agentic AI will autonomously resolve 80 per cent of routine customer service interactions by 2029. Just this month Salesforce deployed specialised AI agent teams into sales workflows, RingCentral launched its AIR Pro agentic platform, and NICE rolled out agentic AI across CXone and Cognigy. Nvidia, meanwhile, continues to power a growing number of autonomous CX agent deployments through its Agent Toolkit.
What distinguishes Oracle’s approach, at least architecturally, is the depth of that native integration. Running agents inside the transactional system rather than layered over existing workflows may allow for faster execution and tighter governance, but how that advantage plays out in practice will partly depend on how organisations choose to deploy the tools.
