TL;DR — Key Takeaways
AI readiness starts with IT fundamentals: ManageEngine CEO Rajesh Ganesan identified six priorities for AI adoption: Infrastructure reliability, data quality and sovereignty, operational resilience, power and compute capacity, governance and compliance, and security.
Context matters more than model selection: Ganesan argued that IT teams already have plenty of AI models to choose from, but lack the operational intelligence, shared data fabrics and orchestration capabilities required to automate workflows effectively.
Premature AI agent deployments increase risk: Without identity controls, security guardrails and appropriate governance, autonomous AI agents could access sensitive information or execute destructive actions, including deleting databases.
The CEO of ManageEngine told hundreds of IT professionals this week that more time and effort need to be applied to getting ready to successfully apply artificial intelligence (AI) to IT operations.
Speaking at the ManageEngine User Conference, Rajesh Ganesan told conference attendees that rather than randomly deploying AI tools and agents that are not going to deliver on the value promised because not enough context is being provided to successfully automate workflows.
In fact, six specific areas that IT teams should be focusing on to make their IT environments ready for AI include upgrading infrastructure to make it more reliable, improving data quality and sovereignty, strengthening operational resilience, addressing existing power and compute constraints, bolstering governance and compliance capabilities and, finally, ensuring IT environments are secure.
In fact, security can not be compromised in the name of productivity, said Ganesan.
At the moment, most IT teams are stuck somewhere between being in an IT Curious phase marked by experimentation and an AI Ready phase, he added. Hardly any have moved into an AI-driven phase where IT operations are managed autonomously by AI agents, noted Ganesan.
The primary challenge today is not determining which AI models to adopt but rather surfacing operational intelligence in a way that can be consumed by AI, he added. “The model is no longer the bottleneck,” said Ganesan. “We are spoiled for choice.”
ManageEngine is currently working to expand the number of AI agents it provides across a portfolio of IT operations tools and platforms that all share access to a common data fabric, noted Ganesan.
Additionally, AI agents will need access to an identity trust fabric that provides the foundation for the governance and security capabilities that IT teams will require to manage and secure them, said Ganesan. In addition, those AI agents will need to be able to access a common workflow and orchestration engine, he added.
Ultimately, the goal is to create a system of workflows through which AI agents inform IT teams about what is about to happen and what actions should be executed to prevent any disruptions, said Ganesan.
The rate at which IT teams are applying AI to IT operations varies widely, but it’s apparent that at the dawn of the agentic AI era there is still much work to be done on IT management fundamentals before they can be trusted to autonomously perform tasks. In general, IT teams will need to make significant investments in a wide range of platforms and capabilities to ensure that AI agents can be safely deployed. AI agents are typically programmed to accomplish a task by any and all means available. The issue is that they are prone to access sensitive data without permission or, in some cases, perform actions such as deleting a database.
The challenge, of course, is that despite those concerns, more than a few IT teams will prematurely deploy AI agents and other related technologies in production environments without first making certain the right guardrails and controls are in place; in the absence of, it will lead to negative outcomes at this point that are all too predictable.
