Thank you to everyone who joined Oracle Integration at Oracle AI World 2026 in Las Vegas. Across sold-out sessions, a hands-on lab, roundtables, executive briefings, and conversations at the Demo Hub, we explored how to put agentic AI to work across the enterprise.

Our message throughout the event was clear: Agents bring the intelligence. Oracle Integration brings the enterprise.

AI agents need access to business applications and data, reliable ways to execute tasks, and controls that keep people accountable for important decisions. Oracle Integration brings connectivity, orchestration, and governance together to help teams move from agent experiments to enterprise business processes.

The Oracle Integration team at Oracle AI World 2026

The Oracle Integration team at Oracle AI World 2026. Thank you to our customers, partners, speakers, and community for making the event a success.

Oracle Integration at AI World in numbers

All four sessions and the hands-on lab sold out, with 672 total attendances across the five activities. Six guest customer speakers brought practical experience to the stage.

The conversations continued throughout the event:

  • 700 webcast attendees joined live from the AI World studio.
  • 250 conversations took place at the demo booth.
  • Three roundtables, including two customer roundtables and one partner roundtable, created opportunities to exchange experiences and priorities.
  • More than 30 executive briefings connected the team with customer leaders.

Session highlights

Building Trusted Agentic AI with Oracle Integration

Building Trusted Agentic AI with Oracle Integration — LRN2138 explored the foundation for trusted enterprise AI: securely connecting agents to authorized applications, orchestrating their actions, and maintaining visibility and administrative control.

The session connected the Oracle Integration product vision with integration, process automation, RPA, observability, connectivity, and AI-assisted development. A central takeaway was the opportunity to reuse existing integrations and business logic as tools for agents, extending investments teams already operate.

[LRN2138] Building Trusted Agentic AI with Oracle IntegrationDownload

Best Practices for Oracle Integration AI-Enabled Projects

Best Practices for Oracle Integration AI-Enabled Projects — LRN2140 focused on organizing development and deployment around business solutions.

Practical guidance covered reusable components, team collaboration, versioned releases, and consistent promotion through development, test, and production. For agent-enabled projects, the discussion emphasized carefully selecting which integrations to expose as Model Context Protocol (MCP) tools, defining clear inputs, and enforcing authorization and business rules in the execution path.

The takeaway: build reusable capabilities, expose tools deliberately, and make every release operationally ready.

LRN2140 – Best Practices for Oracle Integration AI-Enabled ProjectsDownload

AI Agent Governance and Human in the Loop

AI Agent Governance and Human in the Loop with Oracle Integration — LRN2288 addressed a question every enterprise adopting agents must answer: who controls the actions an agent can take?

The session explored governed tool access, approvals, reviews, and human oversight. MCP Gateway and Human-in-the-Loop patterns illustrated how organizations can place controls around agent actions and route consequential decisions to accountable people.

The principle applies across agent platforms: define access, approval requirements, and auditability as part of the business process.

LRN2288_AI_Agent_Governance_and_HITL_with_Oracle_Integration_AIW_3Download

Connected AI Agents with Fusion AI Agent Studio and Oracle Integration

Connected AI Agents with Fusion AI Agent Studio and Oracle Integration — LRN2141 brought connectivity and orchestration into focus through technical discussions and partner perspectives.

An invoice exception scenario illustrated how agents, integrations, enterprise applications, and a human approver can work together. The session explored both directions of interaction: agents invoking integrations as tools, and integrations invoking agents within a larger workflow.

The takeaway: combine agent reasoning with defined business operations, validation, and human approval.

LRN2141 Connected AI Agents with Fusion AI Agent Studio and Oracle Integration AI World 10.2026Download

Hands-on learning with enterprise data and human oversight

Building Agentic AI for SaaS Connected to Enterprise Data with Human Oversight — HOL2143 gave attendees an opportunity to apply the concepts.

The lab focused on designing and orchestrating agents, connecting them to enterprise applications and data, and incorporating Human-in-the-Loop controls into business automation.

Access the LiveLab training here.

Continue your Oracle Integration journey

Bring the learning back to your team by choosing one business process with a clear outcome. Identify the applications and data it needs, select reusable integrations, and define where validation and human approval belong. Use that process to establish a pattern you can extend to additional use cases.

Ready to take the next step? Visit Get started with Oracle Integration for resources to continue learning and building.

Thank you for the ideas, questions, demonstrations, and conversations at Oracle AI World 2026. We look forward to seeing what you build next with Oracle Integration!