“Strategize before you execute, and after you execute, watch and revise.” That principle sits at the core of how Herbert Roy George, Senior Director at DoingERP.com, approaches services managed by artificial intelligence (AI). As enterprises accelerate ERP transformation and cloud modernization, success increasingly comes down to the operational framework supporting AI.
Enterprise resource planning systems, commonly known as ERP, serve as the digital backbone of large organizations, integrating finance, human capital management (HCM), supply chains, and operational data into a single system of record. These platforms are designed to standardize processes and enable real-time decision-making at scale, making them a critical foundation for any AI-driven transformation.
“If you don’t prepare for AI, you can end up automating anything…and not enable it comprehensively enough for the whole thing to work together.” For George, the future of AI-managed business services depends on disciplined architecture, not experimentation alone.
Building the Operational Framework for AI-Driven Enterprise Systems
At the foundation of AI-managed services lies a structured approach to operational design. George points to Harvard’s AGENT framework, which emphasizes audit, gauge, execute, navigate, and transform as the backbone of sustainable AI deployment.
Before implementation, organizations must audit potential workflows and gauge which components are viable for AI. This preparation stage is often overlooked in ERP transformation and Oracle modernization initiatives, where urgency can overshadow more strategic thinking. “You’ll have to figure out which aspects of that particular workflow are good for AI and what should be left as it is.”
Within ERP environments, continuous monitoring and refinement ensure that interconnected systems across finance, HR, and operations remain aligned as business needs evolve. Cloud governance is more about sustaining performance over the full ERP lifecycle, reflecting a broader shift toward lifecycle-based operational intelligence.
From Fragmented Automation to AI-Managed Operations
The gap between task automation and true AI-managed services remains significant. While many organizations believe they are advancing, George outlines a three-stage maturity model: assisting, augmenting, and autonomous.
Most enterprises remain in the assisting phase, where AI supports data retrieval and summarization. The augmenting phase introduces decision support, but still requires human intervention. Only in the autonomous stage do AI systems run end-to-end processes, learning and adapting independently.
“Most businesses are automating tasks but they’re not letting the full cycle run,” George says. This partial adoption creates operational bottlenecks, particularly in enterprise ERP environments where human intervention slows system-wide performance. In AI-driven operations, “the human is going to be the bottleneck,” he adds.
This distinction defines what separates managed services from managed outcomes. True AI-managed services orchestrate entire workflows, from ERP implementation to operational intelligence, rather than optimizing isolated tasks.
Governance as the Backbone of Enterprise AI
As AI systems take on greater responsibility, governance structures must evolve accordingly. George frames AI not as a tool, but as a participant in operations. “You’ve got to think of it just like another employee. You’ve got to train it, you’ve got to feed it with appropriate changes.”
This perspective reshapes how organizations approach cloud governance, HCM architecture, and enterprise scalability. AI systems must be continuously updated to reflect changes in data, processes, and business rules. Without this, even advanced systems risk misalignment with enterprise objectives.
Critical to this governance model is ongoing monitoring. Leaders must evaluate whether AI systems are achieving intended outcomes, whether self-healing capabilities are functioning, and how changes impact performance over time. The navigate and transform stages of the AGENT framework become especially relevant here, ensuring that AI systems evolve alongside the business.
Preparing for the Next Phase of AI-Managed Business Services
The shift toward AI-managed services introduces operational risks that extend beyond technology. Workforce readiness, operating models, and customer experience all require recalibration. One of the most significant changes is cognitive. “AI is bidirectional. It can ask you back questions,” George says, highlighting a departure from traditional software systems. Organizations must retrain employees to interact with AI in more dynamic ways, particularly within talent systems and global enterprise environments. George also emphasizes the need for manual fallbacks. “If something does not go as intended, roll it back.” This mindset reflects a maturing approach to managed operations, where flexibility is as critical as automation.
Customer experience presents another defining challenge. Drawing parallels to legacy interactive voice response (IVR) systems, George warns that poorly designed AI interactions can frustrate users. However, he also sees opportunity. AI systems can respond faster, learn from failures, and proactively correct issues. “That is possible at double the speed.”
The Strategic Imperative Behind AI-Managed Services
The future of AI-managed business services will be shaped by how technology leaders architect AI-driven operations across the enterprise. From Oracle Fusion HCM implementation strategy to building scalable talent systems, the emphasis is shifting toward integrated, governed, and continuously evolving ecosystems. AI is not simply enhancing enterprise systems. It is redefining how they are run. Organizations that align operational frameworks, governance, and workforce readiness will move beyond experimentation toward sustained value.
Follow Herbert Roy George on LinkedIn or visit his website. Learn more about DoingERP.com here.
