Over the past year, we’ve witnessed one of the most significant advances in enterprise technology—the rise of agentic AI. Analysts expect that one-third of enterprise software applications will incorporate agentic AI by 2028, up from less than 1% in 2024. These models can generate code, coordinate workflows, and interact independently across systems. What once required months of development can now be prototyped in days.
For leaders responsible for regulated operations, this raises a natural question: Will AI replace enterprise operational software?
In aerospace and defense (A&D) manufacturing, the answer is straightforward. AI will not replace core manufacturing systems. Instead, it will increase their strategic importance.
AI Is Becoming the System of Decision
With the adoption of AI, a new enterprise architecture is emerging across manufacturing organizations, where AI acts as an intelligence layer that overlays existing systems of record. Enterprise Resource Planning (ERP) remains the financial system of record. Product Lifecycle Management (PLM) governs engineering, and Manufacturing Execution Systems (MES) control execution on the shop floor.
AI operates across these systems, serving as a “system of decision” by aggregating data and offering informed recommendations in real time. This represents a significant evolution that will change how leaders access insights and make decisions. However, decision support is not the same as execution authority. A&D manufacturing requires validated processes, robust security, traceability, and pristine data to ensure compliance with audits. The existing ERP, MES, and PLM systems will remain as the sources of truth, with AI simply helping to interpret their data.
Where AI Delivers Layered Value
These A&D manufacturing systems of record collect and store extensive amounts of operational data, including quality metrics, inspection records, supply chain performance indicators, and production processes. In the past, engineers spent much of their valuable time manually reviewing system data to spot inconsistencies or problematic patterns. AI changes this dynamic.
Advanced planning and scheduling
For example, advanced planning and scheduling often rely on periodic forecasts that are based on static inventory and demand assumptions. An AI-enabled decision layer can continuously evaluate production data, supplier variability, and quality trends to suggest real-time adjustments, enhancing responsiveness to market needs.
Quality risk detection
Similarly, in quality management, AI can analyze serialized nonconformance history, effectivity-driven engineering changes, supplier escape patterns, supplier performance trends, and process deviations to identify emerging risks before they lead to program delays. Instead of reacting to defects, organizations can proactively anticipate them. The result is greater agility and improved quality.
Root cause analysis
When issues arise, AI can quickly review production history, engineering revisions, and supplier inputs to identify potential root causes in hours rather than weeks. This frees up engineers’ time and helps get production back on schedule promptly.
In each of these scenarios, AI improves decision-making, but it must function within defined boundaries. It needs to depend on the ERP, PLM, and MES to provide strategic value.
Execution Still Requires Authoritative Truth
In A&D manufacturing, there is no universal “single source of truth.” Each enterprise platform manages its own domain truth. ERP governs financial and supply chain truth. PLM governs engineering and configuration truth. And MES governs execution truth.
These systems of record, or truth, become especially crucial in the highly regulated environment of A&D manufacturing. As AI copilots and intelligent agents continue to improve A&D manufacturing workflows, architecture will determine whether organizations gain value or introduce risks. If the underlying systems are inconsistent or poorly managed, AI amplifies those issues. If they are strong and domain-specific, AI boosts performance.
In regulated A&D environments, effective deployment requires:
- Clear distinction between decision support and execution control.
- Secure integration with certified record systems.
- Preservation of configuration and traceability boundaries.
- Governance models defining where AI can recommend versus where it can enforce.
In other words, AI should guide execution, not replace its guardrails.
The Strategic Imperative for A&D Leaders
As AI adoption accelerates, operational systems of record become more—not less—critical. A&D manufacturing systems, such as iBase-t’s MES, Solumina, will continue to provide governance, enforce compliance, maintain configuration integrity, ensure security, and support traceability, while AI speeds up decision-making, reveals hidden patterns, and boosts visibility.
Together, they enable a more resilient architecture that can respond to supply chain volatility, accelerate program timelines, and maintain certification rigor.
For A&D leaders, the path forward is not about choosing between AI and core manufacturing systems. It is about strengthening the digital backbone of execution while strategically layering intelligence on top. Embrace AI for what it does best: interpreting complexity and accelerating decisions. At the same time, reinforce the systems of record that serve as the foundation of operational truth.
In an industry where precision and accountability shape success, that partnership is essential and is the foundation for the next level of execution excellence.
Learn more about Solumina AI. Watch this video.
