Twenty-five years after 9/11, one of the most remarkable changes in counterterrorism has been the ability to find terrorist leaders with unprecedented speed and precision. Drones, advanced surveillance, big data, and artificial intelligence have made it increasingly difficult for members of terrorist organizations to hide. AI can process enormous amounts of information, identify suspected militants, track their movements, and generate potential targets far faster than human intelligence officers. Yet that technological progress has produced an important paradox: the better we become at finding terrorists, the stronger their incentive becomes to hide among civilians.

The Identification Problem

The Israel-Hamas war illustrates this dilemma. Israel has reportedly employed AI-assisted systems such as the Gospel and Lavender to identify potential targets and suspected Hamas members. Combined with drones and persistent surveillance, such technologies have dramatically increased the speed and scale at which targets can be identified.

However, Hamas has strong incentives to adapt to Israel’s technological superiority. When militants can be detected and targeted almost anywhere on an increasingly transparent battlefield, civilians become a form of protection against technologically superior forces. Tunnels, densely populated neighborhoods, and civilian facilities provide places to conceal their activities.

That creates a serious identification problem. Counterterrorism forces must distinguish militants from civilians, but violent extremists deliberately make that distinction increasingly difficult. Paradoxically, therefore, improvements in surveillance and identification do not necessarily translate into better civilian protection. Technological advancement can encourage strategic adaptation that places more civilians at risk.

The Limits of New Tech

This dilemma points to a broader lesson from twenty-five years of counterterrorism efforts—military technology can help solve the technical problem of finding and striking terrorists, but it cannot by itself solve the political problem of defeating terrorism. Counterterrorism ultimately depends on distinguishing militants from civilians and maintaining the trust and cooperation of local populations. Civilian casualties can undermine that cooperation, generate resentment, and provide extremist organizations with opportunities for recruitment and mobilization.

The mistake, therefore, is not in employing rapidly evolving technological tools in the fight against terrorism; it is the assumption that these tools produce better strategic outcomes. Twenty-five years after 9/11, we have become extraordinarily good at finding terrorists. But finding terrorists and defeating terrorism are not the same thing. As AI transforms counterterrorism, protecting the civilians among whom terrorists seek refuge may become more important—and more difficult—than ever.

■ 메모 원문 보기

■ 최현진_경희대학교 정치외교학과 교수. 

■ 담당 및 편집: 임재현_EAI 연구원
    문의: 02 2277 1683 (ext. 209) | [email protected]