As enterprise technology platforms develop their own AI assistants, are we creating a more intelligent operating environment—or simply introducing another layer of fragmentation?
Over the past few years, enterprise IT has invested heavily in bringing different technology domains together. Infrastructure, networking, security, identity, workspace and service management may operate through different platforms, but the problems they solve are rarely isolated.
Now AI is entering almost every one of those domains.
Infrastructure platforms are introducing AI assistants. Security vendors are building intelligent investigation capabilities. Workspace platforms are adding their own AI-driven insights, while IT service management platforms are increasingly using AI to automate support and operational workflows.
Each of these developments has value. But looking at them collectively raises an interesting question: Are we making enterprise IT more intelligent, or are we recreating the same fragmentation with a new generation of AI interfaces?
Several conversations with technology leaders and service providers have made me think that the next significant opportunity may not be another AI assistant. It may be the ability to orchestrate intelligence across the enterprise technology estate.
The problem doesn’t respect technology boundaries
Consider a familiar situation: an employee reports that an application is running slowly.
The cause could be an overloaded application server, network congestion, an authentication issue, a poorly performing virtual desktop or a problem with the application itself.
In a traditional operating model, different teams investigate their respective domains. Monitoring platforms generate alerts, service management coordinates the incident, and specialists work together to identify the underlying problem.
Now imagine that every technology platform has its own AI assistant.
The infrastructure assistant understands server performance. The network assistant analyzes connectivity. The workspace assistant understands the employee’s session, while the security assistant investigates whether a security control is affecting access.
Each assistant may provide useful answers. Yet the enterprise could still be left with the same challenge: someone needs to bring those answers together, establish their relevance and coordinate the response.
Intelligence within individual platforms doesn’t automatically translate into intelligence across an enterprise.
From individual assistants to AI orchestration
This is where I believe the next phase of enterprise AI becomes interesting.
Instead of expecting one assistant to understand every technology domain, enterprises could increasingly rely on orchestration that coordinates specialized intelligence.
I think of this emerging concept as an Operational AI Control Plane.
This isn’t necessarily another platform that replaces existing operational tools. It’s a way of thinking about how intelligence from different technology domains could work together to address an operational problem.
The distinction matters.
An infrastructure platform will generally understand its own environment more deeply than a generic enterprise assistant. The same applies to security, identity, networking and digital workspace technologies.
The objective shouldn’t be to recreate all that expertise in one enormous AI system. It should be to make specialized intelligence useful beyond the boundaries of the platforms in which it originates.
An enterprise problem becomes the starting point, rather than the individual technology console.
Domain expertise becomes more valuable, not less
One concern surrounding AI orchestration is whether specialized technology expertise becomes less relevant when a central system coordinates operations.
I see the opposite possibility.
The effectiveness of orchestration depends on the quality of the intelligence available to it. A general-purpose AI system cannot automatically reproduce the depth of understanding that specialized platforms have developed over years of operating in their respective domains.
Take digital workspace as an example.
A user reporting poor application performance might be experiencing an issue with their virtual desktop, application delivery, endpoint or network connection. A workspace platform such as Citrix can contribute domain-specific context that a general infrastructure monitoring system may not possess.
But that context becomes more valuable when it can contribute to understanding the broader operational problem.
The same principle applies to other technology domains.
The future may belong neither to a single assistant that attempts to know everything nor to dozens of assistants operating independently, but to an environment where specialized intelligence works together.
The business case is bigger than AI
For enterprise technology leaders, the value of orchestration cannot simply be the ability to connect multiple AI assistants.
It needs to improve how IT operates.
An employee doesn’t care whether an incident involves infrastructure, networking or workspace. They want their application to work.
Similarly, an IT leader is concerned with how quickly problems are resolved, how effectively specialists are utilized and how much operational effort is required to maintain reliable services.
Cross-domain orchestration could help reduce repetitive investigations, improve coordination between technology teams and shorten the journey from identifying a problem to resolving it.
It could also change how operational expertise is consumed. Rather than requiring every support interaction to move through several teams, an orchestrated approach could make specialist knowledge available at the point where it’s needed.
Of course, coordinating intelligence isn’t the same as authorizing autonomous action. Enterprises will still need appropriate controls, accountability and human oversight, particularly for changes that could affect critical services.
The business objective is not to deploy more AI. It’s to reduce the operational effort required to achieve a reliable outcome.
A different opportunity for GSIs and service providers
This evolution could be particularly relevant to global system integrators and managed service providers.
Unlike individual technology vendors, large service providers often operate across several enterprise technology domains. They understand the operational dependencies between infrastructure, networking, security, workplace and service management because they are responsible for delivering services across those boundaries.
That cross-domain position could become increasingly important as enterprises move from experimenting with individual AI capabilities toward orchestrating them.
The opportunity is not necessarily to build yet another assistant. It’s to turn existing operational expertise and specialized intelligence into more coordinated, scalable managed services.
For technology vendors, this raises a different strategic question.
As orchestration becomes more important, will every vendor need to own the experience through which enterprises consume AI? Or will some create greater value by making their specialized intelligence indispensable to a broader operational ecosystem?
Do you need to own the AI orchestration layer—or do you need to become indispensable to it?
I suspect the answer will differ depending on a company’s position in the enterprise technology ecosystem.
Final thought
The first phase of enterprise AI has largely focused on introducing intelligence into individual applications and technology platforms.
The next phase could be about connecting that intelligence so it can address problems that extend beyond individual products and operational teams.
For enterprises, the question will increasingly shift from how many AI assistants they have deployed to how effectively those capabilities work together.
For technology vendors and service providers, it creates an opportunity to reconsider where they contribute the most value.
The next significant improvement in IT operations may not come from making every individual assistant smarter. It may come from making the enterprise’s collective intelligence work better together.