Global rivalries now unfold more through influence operations, narrative battles, and incremental pressure than through direct clashes on conventional battlefields. Russian actions in Ukraine, Iranian proxy activities across the Middle East, and Chinese maneuvers in the Indo-Pacific all demonstrate how adversaries blend information tools with local networks to shape outcomes below the threshold of open war. American leaders have access to vast intelligence flows and analytic resources, yet persistent gaps in how those inputs translate into reliable strategic choices continue to surface. Past experiences, especially the long engagement in Afghanistan, offer clear warnings about what happens when decision systems prioritize confident-sounding conclusions over calibrated judgment.
Studies of national security professionals reveal a consistent pattern of miscalibrated confidence. When experts assign 80 or 90 percent certainty to their assessments in complex political environments, actual accuracy often lands closer to 60 percent. This gap arises less from individual analysts than from institutional processes that aggregate information poorly, rarely track past performance against real outcomes, and present polished briefs that mask underlying uncertainty. In fast-moving situations where informal networks and social trust matter most, these habits create blind spots that kinetic superiority cannot easily fix.
Afghanistan’s Lasting Lessons
Few operations generated as much internal review as the two-decade effort in Afghanistan. Afghanistan reviews and inspector general reports repeatedly identified the same core problems: overestimation of partner capacity, failure to grasp how corruption and predatory governance undermined legitimacy, and persistent optimism that ignored ground-level warnings. Yet these findings rarely fed into a unified system that updated assumptions across agencies or carried hard-earned insights into subsequent planning cycles. Each new strategy often restarted with fresh assumptions rather than building on accumulated evidence.
The result was not sudden collapse in 2021 but a slow structural failure prepared over years. Without mechanisms for shared data models, explicit testing of recurring hypotheses, or accountability for repeated analytical errors, knowledge accumulated in archives instead of shaping future behavior. Similar dynamics appear when Washington confronts other complex environments today. Adversaries exploit exactly these lags, maintaining influence structures that survive leadership changes while American approaches reset with each administration.
Domestic initiatives have mirrored these tendencies. Recent efficiency reforms focused intensely on measurable targets such as staff reductions and immediate budget savings. Less attention went to second-order effects on specialized oversight, crisis response capacity, or the hidden dependencies that sustain government resilience. The parallel is instructive: optimizing for what analysts can easily count can obscure the factors that actually determine long-term outcomes in both foreign and domestic arenas.
Building Genuine Decision Infrastructure
Environments characterized by gray zone competition punish such miscalibration with special severity. Opponents avoid decisive engagements and instead chip away at legitimacy, elite cohesion, and public confidence over time. By the moment problems register on conventional metrics, decisive ground has often already been lost. Research on cognitive security shows how targeted influence can degrade decision quality before traditional warning signs appear.
The United States does not lack talent or technology. What it needs is connective architecture—interoperable systems that log assumptions, measure forecast accuracy over time, and force regular confrontation with contradictory evidence. Forecasting research demonstrates that structured techniques, including explicit probability ranges and rapid feedback loops, can markedly improve reliability on bounded questions. High-reliability organizations in other fields have shown how continuous learning prevents the same mistakes from compounding across cycles.
Policy lessons from past engagements remain valuable only if they drive genuine organizational change rather than another round of reports. As hybrid pressures intensify worldwide, investing in these learning systems offers the clearest path to turning data abundance into durable strategic advantage. Without that shift, even well-intentioned policies risk repeating familiar patterns of initial assurance followed by costly correction. The opportunity to close this gap exists. The coming years will test whether institutions seize it.
Original analysis inspired by Masoud Andarabi from The Cipher Brief. Additional research and verification conducted through multiple sources.
