AI Pushes Enterprise Transformation Beyond IT as Agentic Systems Reshape How Companies Operate — DataArt - September 28, 2026
Artificial intelligence is changing the mechanics of enterprise transformation as organizations give employees greater ability to build applications, automate work and redesign processes without relying exclusively on centralized IT departments. The shift is forcing CIOs and other senior executives to rethink governance, organizational structures and leadership practices as technology creation spreads across the enterprise.
The implications extend beyond another cycle of technology modernization. Generative AI, agentic systems and increasingly accessible development tools are reducing the technical barriers that once separated technology producers from technology users. Employees who previously depended on IT teams to develop applications or automate processes can increasingly create their own AI skills, workflows and software.
That democratization could accelerate innovation, but it also creates a management challenge. Enterprises must find ways to encourage experimentation while maintaining security, quality, architectural discipline and alignment with business priorities.
During a BizTechReports executive vidcast interview, Yuri Gubin, Chief Technology Officer at DataArt, said that these forces are contributing to a larger change in how companies approach transformation. Rather than treating transformation as an initiative with a defined beginning and end, organizations increasingly need mechanisms for continuously identifying technological change, evaluating its implications and adjusting business operations. This, he added, puts pressure on executives to develop an organizational capacity for transformation rather than simply managing individual transformation programs.
AI decentralizes technology creation and raises the governance stakes
Enterprise IT has been moving toward greater decentralization for years as cloud infrastructure services, software-as-a-service applications, low-code platforms and consumer technologies gave business units greater control over technology. Generative AI is significantly accelerating that trend.
Gubin described an environment in which the traditional distinction between IT and the rest of the business becomes increasingly difficult to maintain. Employees can use AI to create automations, build software and develop specialized capabilities regardless of whether they possess conventional programming skills.
The expansion of development capacity changes where technology risk originates and how it must be managed. As application creation spreads across business functions, organizations can no longer rely exclusively on traditional IT development processes to enforce architectural standards, security controls and operational discipline.
The resulting challenge is less about protecting software development as a specialized function and more about establishing an enterprise environment in which decentralized innovation can occur safely, changing the role of technology professionals. Architecture, governance and platform engineering become increasingly important because organizations need common environments through which employees can experiment without creating unacceptable security, compliance or operational risks.
The model effectively shifts some of IT’s responsibility from building individual applications to creating the enterprise environment in which applications can be safely built. As AI gives employees across finance, operations, marketing and other functions greater ability to develop their own tools and automations, IT increasingly becomes responsible for providing the common architecture that makes a distributed development manageable. Rather than developing every solution, technology organizations can establish platforms, standards and guardrails that allow other employees to create solutions themselves.
Companies are already experimenting with organizational structures designed to support that approach. AI centers of excellence, cross-functional committees, competency centers and AI enablement teams can provide horizontal governance across organizations that historically managed technology through vertically organized business units.
Gubin said those groups can identify successful practices, resolve bottlenecks and help different departments learn from one another. Importantly, they need participation beyond IT. Representatives from business functions can act as AI champions or ambassadors, bringing operational knowledge into decisions about how AI is deployed.
Without those mechanisms, decentralized experimentation can produce duplication and waste as separate teams independently discover the same lessons or repeat the same mistakes.
Leadership shifts from setting transformation goals to building capacity
The decentralization of technology also changes what senior leadership must contribute to transformation. Executive sponsorship has traditionally meant setting strategic priorities, approving investments and holding managers accountable for results. AI adoption requires greater technological fluency because executives increasingly must make decisions about capabilities that are changing faster than conventional planning cycles.
Gubin said executives do not need to become technologists, but they do need sufficient AI vocabulary and a working understanding of the technology to evaluate opportunities, establish meaningful targets, allocate resources and determine how roles and learning programs should evolve.
That capability becomes particularly important because AI can change organizational capacity itself. A department that could automate only a limited portion of its workload two years ago may now have substantially different options. Leaders therefore need ways to observe how AI is affecting work across the organization rather than relying exclusively on traditional technology performance measures.
That could require a broader definition of observability. Technical telemetry remains important, but executives may also need information about where business processes are struggling, which groups are adopting AI, where skills are developing and whether employees have received appropriate training. Those indicators can help management understand whether the organization is developing the capabilities required to meet its transformation objectives.
The information also needs to travel upward. Employees closest to business processes often understand operational bottlenecks better than senior executives. They can identify repetitive work, inefficient handoffs and opportunities for automation that may be difficult to detect from the executive suite. Hierarchical reporting structures, functional silos and limited visibility across departments can prevent those insights from reaching leaders with the authority and resources to act on them.
Gubin argued that organizations therefore need both top-down direction and bottom-up discovery. Leadership establishes priorities and boundaries while employees identify practical opportunities for applying AI.
Connecting those two directions requires deliberate knowledge sharing. Forums, communities and cross-functional structures can give employees a place to demonstrate successful experiments and allow other departments to reuse what has been learned.
That reduces what Gubin described as information asymmetry, which occurs when employees, departments and executives have different levels of visibility into what is happening across the business, limiting their ability to make informed, coordinated decisions. Without mechanisms to address this, different layers and departments can interpret transformation objectives differently, limiting an enterprise's ability to scale successful innovations.
Culture becomes infrastructure for continuous transformation
The increasing speed of technological change also raises the importance of corporate culture. Transformation programs often focus heavily on technology architectures, financial objectives and implementation schedules. Agentic systems introduce a more persistent organizational challenge because employees must repeatedly reconsider how work is performed as capabilities evolve.
Culture can either accelerate that process or create resistance. Gubin said organizations undergoing significant change need clearly articulated principles that employees can understand without layers of explanation. Those principles help employees make decisions when technology and operating conditions are changing faster than formal policies can be rewritten.
Customer expectations provide another reference point. Transformation ultimately has to connect internal technological capabilities with changes in how customers perceive value and what they expect organizations to deliver.
The same discipline applies to AI experimentation. Proofs of concept have proliferated as companies investigate generative AI, but technical success does not necessarily translate into business impact. Gubin said DataArt has encountered situations in which clients initially requested proofs of concept but benefited from first examining what business outcome the proposed project was supposed to produce.
A technically successful experiment can create only limited value if the organization has not established how it will affect operations, revenue, costs or customer outcomes. That distinction is becoming more important as companies move beyond the initial AI experimentation cycle. Enterprises have accumulated enough experience with generative AI to recognize that producing working prototypes is considerably easier than changing business performance.
DataArt has applied some of the same principles internally, using its own AI adoption efforts to examine governance structures and centers of excellence before applying those experiences to client engagements, according to Gubin.
The larger market lesson extends beyond any individual technology provider. AI is lowering the barriers to creating technology at precisely the moment when enterprises are becoming more dependent on technology for competitive differentiation.
That combination makes organizational capacity a central component of transformation. Companies that can distribute innovation while maintaining governance, share knowledge across organizational boundaries and continually reassess what their workforce can accomplish may be positioned to adapt more quickly as AI capabilities advance. Gubin concluded that those who treat AI primarily as another technology implementation risk missing the larger shift taking place inside the enterprise.
###