Nick Burling, Chief Product Officer at Nasuni.
We’ve been asking the wrong question about enterprise infrastructure.
Enterprise IT has spent years optimizing where workloads should run. Should workloads run in the cloud or on-premises? Should architecture be centralized or distributed? Those decisions still matter, but AI expands the conversation by asking where work actually happens.
The prevailing strategy was simple: Pull everything toward the center. Applications moved to the cloud, and data followed because centralization solved problems that every IT team was wrestling with, from securing enterprise data to managing increasingly complex infrastructure.
That approach assumed people would come to the data. Increasingly, the opposite is true. Employees collaborate across locations; business operations generate data far beyond the traditional office; and AI is expected to operate wherever that work is happening.
As AI becomes embedded across the enterprise, architecture is increasingly defined by how securely data can be accessed wherever work happens. That’s where the edge comes into play.
Why does the edge matter again?
Hybrid work changed the enterprise in ways that are still unfolding. AI arrived just as organizations were learning to operate across offices, homes, job sites and every environment in between, placing new demands on enterprise architecture.
Organizations now generate valuable operational data far beyond traditional office environments.
For example, an engineer reviewing a multi-gigabyte design file from home faces many of the same architectural requirements as a team working in the field. Both depend on timely access to trusted enterprise data. Too often, those employees are still relying on infrastructure built for a different era of work. Large design files, engineering datasets and media assets place demands on enterprise networks that traditional remote access approaches were never intended to support.
As distributed work becomes a permanent part of enterprise operations, architecture has to evolve alongside it. That’s why the edge deserves a second look.
One of the biggest misconceptions I hear is that the edge is a specific place. People picture a factory floor, a retail store or an IoT deployment. Those environments are certainly part of the edge, but they represent only a fraction of where enterprise work happens today.
AI is most effective when it can understand the operational context surrounding a decision. The farther data has to travel or the more times it has to be copied before AI can use it, the harder it becomes to preserve that context. The edge, therefore, becomes more than another deployment location. It becomes an extension of the enterprise where intelligence increasingly needs to operate.
AI changes the role of the edge in enterprise architecture. Treating it as part of the data strategy creates a stronger foundation for scaling AI.
AI depends on a consistent data foundation.
The edge changes the economics of data movement. As organizations generate and use data across more environments, moving massive datasets back to a central location becomes an increasingly expensive way to deliver intelligence. AI performs best when it can work from data in its operational context, making constant movement less practical as AI expands across the business.
That changes how enterprise architecture approaches data access. Every new workflow creates the temptation to duplicate data so people or AI can reach it more easily. Over time, those copies drift apart, making the environment harder to govern and reducing confidence that everyone’s working from the same information.
The architectural objective is becoming much simpler—make trusted enterprise data available wherever work happens instead of creating another copy to support every new location or application. Extending secure access preserves a consistent data foundation while allowing people and AI to work from the same information.
That foundation gives organizations the flexibility to adopt new AI capabilities without having to rethink the architecture at every step. It also allows enterprise architecture to evolve alongside AI as work continues to become more distributed.
Why are enterprises cautious about changing the model?
Enterprises have been slow to move away from architectures they understand. Years of investment have embedded applications and permissions into environments, making the status quo feel safer than architectural change. Even when the long-term case for modernization is compelling, near-term disruption can be enough to delay action.
The risk of changing architecture is easy to see, but the cost of leaving it unchanged emerges more slowly. Concerns around migration, cost and maintaining security through transition reinforce the instinct to stay put, even as existing environments become more expensive or difficult to adapt.
The industry has a responsibility to make that decision easier. Enterprises need a predictable path to modernize incrementally while preserving the controls they trust. Greater flexibility becomes easier to embrace when organizations can evolve their architecture without introducing a new category of operational risk.
Making the transition predictable is essential as distributed data access becomes more important to enterprise AI.
The new enterprise architecture is everywhere.
When organizations can rely on a consistent data foundation, new AI capabilities can be adopted with relatively little friction. When those foundations are fragmented, every new initiative begins by solving infrastructure problems before it can deliver business value.
That’s what changes the architecture conversation. Infrastructure stops being the destination and becomes the enabler. Architecture decisions increasingly influence the pace of innovation just as much as operational efficiency because they determine how easily new AI capabilities can be introduced as business needs evolve.
The rise of agentic AI raises the stakes even further. AI agents interact with enterprise data much like employees do. They depend on secure, governed access to enterprise data. Architectures built around moving or duplicating data create unnecessary friction for both.
Modernizing enterprise architecture increasingly means modernizing how enterprise data is accessed. A trusted, governed data foundation allows employees and AI to work from the same information while giving organizations the flexibility to adapt as enterprise AI continues to evolve.
The edge has become part of that equation because work has changed. It reflects a broader shift in enterprise architecture, where extending secure access to trusted data has become just as important as deciding where infrastructure runs.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
