AI appears to be creating increasingly interconnected enterprise ecosystems and expanding the complexity of how organizations manage technology throughout their operations. As artificial intelligence penetrates deeper into critical workflows, maintaining visibility into system dependencies is emerging as an important issue for leadership. according to AI sovereignty research91% of managers who participated in the survey said that they do not fully understand their organization. AI dependencies. Meanwhile, respondents also reported an average of six AI-related breaches in the previous two years. Together, these findings suggest that management practices may require evolution along with artificial intelligence.
Jeffrey Rachlin and partner Andy Hyman observed a similar pattern in complex environments. In practice, many organizations continue to investigate failures after an apparent breach has occurred. As AI systems assume greater autonomy retrospective analysis across business processes can offer only part of the picture and provides an opportunity to consider management techniques that identify meaningful changes where intervention is possible.
Jeffrey Rachlin
This perspective reflects a broader shift in how organizations can think operational health. Monitoring often highlights results through dashboards, reports and key performance indicators. The duo emphasizes that these tools remain valuable, but they typically describe the outcomes generated by the system rather than the relationships within the system that generate those outcomes.
When performance indicators show concern, the conditions that contributed to that result may have been developing for some time. Hyman and Rachlin believe that organizations can benefit from supplementing performance monitoring with a greater focus on system behavior, interaction patterns, and evolving dependencies that affect resilience long before visible disruptions occur.
💜 of EU technology
The latest rumblings from the EU tech scene, a story from our wise founder Boris and some questionable AI art. Free in your inbox every week. Register now!
Rachlin explains:Continuity begins to fail long before the disruption becomes apparent. When organizations develop the ability to understand how their systems change, they strengthen their future when those changes are still manageable.“
This philosophy aligns with Hyman’s Marginal Point of Systemic Drift (MPOSD) framework, which suggests that specific patterns may indicate that management visibility is declining before operational results are revealed. Rather than attempting to predict every future event, the framework focuses on identifying structural signals that may indicate that the system is becoming increasingly difficult to evaluate independently.
Rachlin and Hyman identified five recurring indicators that appear together in many complex system scenarios. First, verification integrity violations reflect situations where system results evolve faster than independent verification processes. When alerts, reviews, or operational metrics no longer provide an accurate representation of system activity, proxy override escalation occurs.
Andy Hyman
Stimulus mismatch describes situations where a system has limited structural stimulus to detect its drift. As delays between action and visibility become increasingly meaningful to decision makers, delay inflation and feedback distortion emerge. Finally, an erosion of management independence develops when control mechanisms rely on the same systems designed to evaluate them.
According to the duo’s observations, these signals are especially meaningful when combined rather than seen in isolation. Hyman says:Complex systems are rarely difficult to control at once. Governance changes as independent visibility begins to diminish, and embracing this transition can provide valuable opportunities for informed decision-making.“
According to Rachlin, the importance of independent visibility has become easier to appreciate through recent AI events. In a casethe autonomous encryption agent deleted production data and backups within seconds after operating outside of its intended limits. Hyman and Rachlin’s retrospective application of the MPOSD suggested that observable indicators may emerge before the irreversible stage of the sequence. Although retrospective analysis cannot determine future outcomes, the duo believes this case demonstrates how earlier identification of structural changes can broaden the range of management decisions available before a disruption occurs.
This perspective is intended to encourage leaders to rethink how organizational health is measured. Dashboards and KPIs remain meaningful components of performance monitoring, but increasingly interconnected AI ecosystems can also benefit from monitoring the relationships that connect systems. An independent assessment of management health, viewed separately from the systems being evaluated, can provide additional context that supports more informed operational decisions as complexity continues to increase.
Rachel says:AI will likely continue to increase its presence enterprise settingsopens up new opportunities, but also raises new questions about how organizations manage and direct its use. Technology can offer powerful opportunities, but a company’s ability to remain resilient may also depend on spotting changes early before they become larger operational problems.“
As Hyman and Rachlin’s work shows, anticipating systemic drift can complement traditional management in ways that support more informed leadership decisions. In addition to thoughtfully responding to visible results, organizations that continue to develop the ability to recognize early signals can help define the next chapter of innovation with greater confidence and resilience.
