Ansible Automation Orchestrator: Unified AI-Driven IT Operations Workflow

The Automation Orchestrator is a new unified experience for AI-driven IT operations — coming Q3 2026. It connects event detection, AI reasoning, and deterministic execution in a single, auditable pipeline.

Core Principles

• ✅ AI recommends — AI analyzes events and suggests actions • ✅ Humans approve — approval gates ensure oversight • ✅ Automation executes — deterministic execution through AAP

> "AI isn't improvising against production infrastructure, it's acting through AAP."

See also: AIOps: The Future of IT Operations with AI-Driven Automation

What It Does

One governed workflow, end-to-end: • Connects event detection, AI reasoning, and deterministic execution in a single auditable pipeline • Seamless orchestration of disparate tools and processes • Visual workflow builder for multi-step automation chains

Example: Vulnerability Remediation Pipeline

The Orchestrator UI (at ) provides a visual canvas where agents, events, and playbooks are orchestrated on a single screen.

5-Step Workflow

Step 1: Alerts from multiple sources Events arrive from IBM Instana, ServiceNow, and Splunk vulnerability alerts — all as Event-Driven Ansible triggers on the same canvas.

Step 2: Events trigger deterministic automation rulebook The alerts feed into a "Create SNOW Ticket" job template that creates tracking tickets and a "Create JIRA Ticket" action — deterministic, auditable steps.

Step 3: AI analyzes and recommends A "Vulnerability Analysis" AI agent step uses a configured model (e.g., GPT-4) with defined tools (4 tools shown) to analyze the vulnerability and recommend remediation.

Step 4: Humans approve A "Human Review" approval gate with timeout and escalation. Reviewers see the AI recommendation and can approve or reject. Green = approved, red = rejected.

Step 5: Automated remediation at scale An "Execute Remediation" AAP job template runs the "Patch and Restart" playbook across affected hosts.

┌─────────────────┐     ┌─────────────────┐     ┌──────────────┐     ┌──────────────┐     ┌──────────────┐
│ 1. ALERTS        │     │ 2. AUTOMATION    │     │ 3. AI AGENT   │     │ 4. HUMAN      │     │ 5. REMEDIATE  │
│                  │     │                  │     │               │     │    REVIEW      │     │               │
│ IBM Instana     │────►│ Create SNOW      │────►│ Vulnerability │────►│ Approval gate │────►│ Execute       │
│ ServiceNow      │     │ Ticket           │     │ Analysis      │     │ (Timeout +    │     │ Remediation   │
│ Splunk          │     │ (Job Template)   │     │ (Model + 4    │     │  Escalation)  │     │ (Patch and    │
│ (EDA Triggers)  │     │                  │     │  tools)       │     │               │     │  Restart)     │
└─────────────────┘     └─────────────────┘     └──────────────┘     └──────────────┘     └──────────────┘

Canvas Legend

| Icon | Step Type | |---|---| | AI Agent | AI-powered analysis/recommendation | | Action | Generic automation action | | AAP Execution | Job template or workflow | | Logic | Conditional branching | | Approval | Human review gate | | Trigger | Event source (EDA) | | Green connector | Approved path | | Red connector | Rejected path |

See also: Ansible AI-Driven Automation Architecture: 4 Entry Points for GenAI, API, EDA, and MCP

Built for Every Automation Persona

| Persona | Interface | |---|---| | Platform engineers | Intuitive GUI-based workflow builder | | IT operators | Visual pipeline monitoring and approval | | Automation developers | Headless API and MCP integrations |

How It Differs from Existing Workflows

| Feature | Current AAP Workflows | Automation Orchestrator | |---|---|---| | AI integration | Manual job chaining | AI recommends next steps | | Event-driven | Separate EDA rulebooks | Unified event → action pipeline | | Approval gates | Workflow approval nodes | Contextual AI + human approval | | Multi-tool | AAP-only execution | Orchestrates external tools (ServiceNow, Splunk, etc.) | | Visibility | Per-job logs | End-to-end pipeline audit trail |

See also: Red Hat Ansible Automation Platform Roadmap 2026: AAP 2.6 to Q4 Timeline

Timeline

| Milestone | Date | |---|---| | Technology Preview | AAP 2.7 (June 2026) | | General Availability | Q3 2026 |

Access the technology preview at .

FAQ

Is the Automation Orchestrator a separate product?

No. It's part of Ansible Automation Platform, accessible through the Ansible Portal.

How does it relate to AAP Workflow Job Templates?

The Orchestrator builds on top of existing workflows but adds AI reasoning, external tool integration, and contextual human approval gates. Existing workflow templates continue to work.

Can I use it without AI?

Yes. The Orchestrator works for traditional deterministic workflows too. AI is optional but adds intelligent routing and recommendations.

What external tools can it orchestrate?

Any tool with an API or webhook: ServiceNow, Splunk, IBM Security, Jira, PagerDuty, and more — connected through the visual workflow builder.

Related Articles

• Ansible AI-Driven Automation Architecture • MCP Server for AAP • What's New in Event-Driven Ansible • Red Hat Ansible Automation Platform 2.7: What's New • Ansible Roadmap