On the final evening of Knowledge 26, Idris Elba stepped behind a set of turntables at the Sphere in Las Vegas. Most people in the audience knew him as an Emmy nominated actor and, more recently, as a ServiceNow spokesperson. Fewer knew he's been DJing in English pubs for years — long before the rest of us caught on. But that night, with the Sphere's massive wraparound display and concert-grade sound at his back, the same craft he'd practiced in small rooms filled an arena. It's hard to forget the experience of your entire body becoming an amplifier for the underlying beats. Idris's skill set hadn't changed. But he was operating at a completely different scale. It was magical.
The visceral experience of that performance was still hanging on me as I flew home that night. I reflected on the dozens of conversations I had that week on the Expo floor — at the workshops where customers got real about technology and data maturity. I met practitioners across a surprisingly wide spectrum. A few were ready to deploy autonomous AI workers across enterprise-level operations. A few were still working to map out their small office networks cleanly. Both groups, and all those in the messy middle, were doing serious, necessary work — the distance between their "arenas" was bigger than I expected — but the underlying skills they brought to the table and the problems they were trying to solve were not that different.
The Q2 store releases are built to track with you wherever you are on that journey. Some of what you'll read about below is available through controlled availability for our most mature customers — early access to the autonomous capabilities we've been building toward. The rest ships broadly, and it addresses the practical work of making IT Operations smoother, faster to stand up, and harder to break. Wherever you are in the journey, there is something magical for you in the 2026 Q2 innovations.
We've assembled four members of our outbound product team to write about what's new across ITOM and the CMDB. Find the areas that matter most to your environment and dive in.
ITOM Visibility |
The gap between buying ITOM and actually using it has been wider than it should be. Derek Kitzelman's blog covers the Q2 capabilities designed to close it: the ITOM Product Hub that puts everything you're entitled to one click away, configuration consoles that guide setup for Discovery and Event Management without needing an expert in the room, URL Discovery that surfaces the SaaS and AI applications your employees are actually using — including the ones nobody approved.
If you've ever suspected there's more running in your environment than your inventory shows, start here.
AIOps and Observability | From spectacle to standard
Routine alerts consume the hours your team should spend on the incidents that actually threaten service. Victoria Lo's blog introduces the AIOps AI Specialist and the SRE AI Agent — autonomous team member components now in controlled availability that handle triage, investigation, and evidence-based root-cause analysis across the observability tools you already run, from Datadog and Dynatrace to Splunk and CloudWatch. When a major incident is resolved, AI agents analyze your CMDB endpoints and stage synthetic monitors to close the coverage gaps the incident exposed.
If your team spends more time assembling context than acting on it, this is the blog to read.
CMDB and Data Foundations | 157 Sources, 1 Truth
A trusted, fit-for-purpose CMDB (not perfect!) is what often separates operations that crawl the pubs from operations that are ready for the spotlight. That's because AI agents need trustworthy data to reason over, and the data flowing into the CMDB from dozens of sources needs to produce a single, clear context.
Jami Tucker's blog covers the Q2 advances that make both easier: an MCP Server that lets AI agents query and interact with the CMDB directly, an AI-assisted de-duplication agent that accelerates remediation with evidence-based rationale, a what-if analysis tool for testing Dynamic IRE rules before committing, and a new Success Advisor that measures CMDB health through the lens of the outcomes the data actually supports.
Your technology data is the foundation that so much else depends on, and Jami's blog explains why you are far more likely to succeed in your CMDB work after Q2 than at any time in the past ten years.
LEAP | Still Solving the Same Issues Twice?
Every incident your team resolves is either a one-time fix or the start of automation that handles the next one. Gian Mario DeLuigi's blog covers Q2 advances to LEAP — the Learning-Enhanced Automation Platform — where an AI agent now matches existing Ansible playbooks to resolution steps based on what's actually worked before.
If no playbook fits, the steps feed directly into a new one. The matching sharpens with every execution, so the work your team has already done compounds over time. If you've been looking for where resolved incidents stop being lost knowledge and start becoming reusable automation, this is that blog.
The conversations I had at Knowledge ranged from teams just getting their arms around what they have to teams ready to hand entire alert workflows to an AI Specialist. Both are real starting points, and both scales just got easier with the features in this release. Just as digital turntables and drum machines opened the music industry to folks like Idris with a couple of LPs and an idea, you can expect that AI agents, MCP Servers, and conversational install assistance will lower the barriers to enterprise-level IT operations.
Click through to the blogs that match your environment and see what's new.
