Commvault SHIFT Virtual 2026 | Recovery is the New Resilience
So the last week has certainly been a turbulent one within the technology industry. I know I was certainly impacted as I decided to get on top of some of my backlog and was soon slamming on the brakes. At first I just assumed it was my connection and within a few minutes I realized that Outlook was the thing effected and this is when the mass of messages with friends and family came through with them having similar issues so it was not just me. Luckily for me I was able to get into most the portals I needed with only a few 2FA codes causing me issues due to time outs. This is then where I was thinking if I am having an issue paying bills and gaining access to customer support how much of an impact can this have on the upstream business and also economy.
It then got me further thinking about why are we seeing this more and its quite simple when you take a step back, so many businesses and ends users rely on platforms such as these when as issue does occur to this scale the blast radius is subsequently just as large. Back when everyone had their own server(s) doing this it was only that business effected but now as its all a shared model it can impact a much larger chain. This is quite often the shadow cost of not running your own infrastructure or being easily able to recover your data.
What we need to remember is that it does not matter whether the cause is an attacker, a bad config push or a third party going dark. What matters is whether you can recover quickly and keep the business running.
I for one always like to try and stay abreast of where all vendors are going especially in the velocity created by AI. We are having to make decisions sometimes at break neck speed where previously we may have had more time. I am finding more and more overlap within the market but it is ultimately we need to find the right tool fort the job at hand.
Resiliency is built up of many components and this is for sure something I am exploring over the coming months once again. Quite often we can not just rely on one solution or architecture and if we do how can we ensure that this can be recovered efficiently. We need to ensure we have solid foundations of infrastructure and data security methods. Also with the age of AI if the same weaknesses are within the recovery how long until they are exploited again.
This week Commvault SHIFT Virtual is one day online on November 18 with keynotes, customer perspectives and expert sessions
on why recovery is the new resilience. It takes all the great content from the Nashville conference global, so anyone who missed SHIFT in person can see the keynotes, announcements and demos.
It is free, broadcast live in four regional slots, and everything is on demand from November 19.
START TIMES, NOVEMBER 18
EMEA 9:00am GMT | India 2:30pm IST | APAC 12:00pm AEDT | AMER 1:00pm ET
Register here to check this out: https://www.commvault.com/shift-virtual
AI is now the operating fabric of the enterprise. More intelligence, more automation, more opportunity, more at stake, all at unprecedented speed. That puts a whole new premium on the ability to recover. SHIFT is where recovery clears the way to test new ideas, move through uncertainty and act on every opportunity AI creates.
There are six key tracks across the day tracks: AI and Data Security, Cyber Recovery, Identity Resilience, Simplify Hybrid Complexity, Customer Stories and Hands-On Labs. Sanjay Mirchandani’s opening keynote, “The AI Recovery Imperative”, sets the direction.
- Dual-Sided AI Risk. Enterprises face threats from both external attackers using AI to accelerate breaches and internal autonomous AI agents making destructive configuration or data errors at machine speed.
- Full-Stack Business Restoration. Restoring isolated files is no longer enough. True resilience requires rebuilding interconnected business services—including applications, identities, network rules, and AI agents—in the exact right sequence. Agents are not human but can also make destructive mistakes and how to we regress back to our original estate
Looking through these sessions there are a few key announcements you can look further into
- Recovery Studio. Moves recovery from restoring data to restoring the business. Runbook-driven orchestration that brings applications, identities, configurations, infrastructure and data back in the right order, with threat scanning and validated checkpoints built into the workflow. Commvault’s own figure: recovery to even minimal operation averages 28 days with conventional approaches. Early access December.
https://www.commvault.com/news/commvault-recovery-studio - AgentRecover, the AI Agent Recovery Plan. Discovers AI agents and their dependencies, audits what they did, and recovers both the agents and the systems they touched when an agent makes changes it should not have. Early access for Microsoft Foundry in November and Amazon Bedrock AgentCore in December.
https://www.commvault.com/news/ai-agent-recovery-plan - Zscaler partnership. AI discovery and policy enforcement from Zscaler connected to recovery of data, applications and identities from Commvault, with planned collaboration on runtime governance and agent detection.
https://www.commvault.com/news/strategic-partnership-with-zscaler - Databricks. Commvault Data Activate for Databricks puts protected, governed enterprise data to work for AI. Early access December in select Azure regions.
https://www.commvault.com/news/commvault-data-activate-for-databricks - NVIDIA. NVIDIA NIM microservices and Nemotron Retriever working with Commvault Data Activate so enterprises can use more of their data for AI safely. Reference design early release December, GA 2027.
https://www.commvault.com/news/work-safely-for-ai-with-nvidia
One other thing that stands out to me amongst all of this is also have to we secure and recovery identity alongside be aware of where our data resides.
Also these items are great but we need a way of proving who we are and if that has gone what effective ways do we have to recover this. Is one identity provider even enough anymore or do we need to ensure that we have full control and auditability of that data source so we can ensure recovery. Especially with agents do they have more access than they necessarily need causing a potential to create damage to your infrastructure and/or data. Quite often identity if now being migrated to a similar shared platform for the connivence but that also comes with a risk of this going down and maybe relying on a third party performing the recovery for you
Another item that I want to explore more is the ability to tag and understand your data foot print. Especially with AI we consume large amounts of data and quite often more than we need and this could also present risk factors. What if that model or agent ingests PII of which it should not have as its so much easier to users to upload a whole file or connect to a database/lake and subsequently what happens if that model then decides to train from that data should it have even had access to this in the first place and if its then part of the model should other users then be able to access this data as it originally those files could have had different access controls.
These thought are my own and I am exploring this further as I want to fully understand the landscape especially with the new avenues AI has created and how these create and also solve business risk.


