Person:

Tzukert, Nimrod

Semi-State Terrorist Organizations (SESTOs) are armed groups that not only control territory but also govern civilian populations, blending militant violence with governance functions. Despite their rarity, SESTOs account for a disproportionate share of terrorist attacks and fatalities globally. Here we identify 24 SESTOs using explicit criteria applied to organizations in the Global Terrorism Database (1970–2020), constructing an organization-quarter panel to compare their violent activity against thousands of non-SESTOs. Fixed-effects regressions reveal that SESTOs sustain significantly higher attack frequencies and aggregate fatalities without systematically increasing per-attack lethality or targeting state actors disproportionately. Notably, SESTOs exhibit episodic, extreme bursts of violence. A Random Forest machine learning model trained solely on violent activity accurately distinguishes SESTOs from non-SESTOs, confirming their distinct operational profile. These findings suggest that governance capacity shapes SESTOs' sustained and intense violence, highlighting their outsized threat and the need for nuanced policy responses that consider their unique behavioral patterns.

Event-level terrorism datasets are designed for incident-level analysis, whereas many research questions require data better suited to organizational trajectories and within-organization change over time. We introduce TOQA (Terrorist Organization Quarterly Activity), a reproducible organization-quarter panel dataset derived from the Global Terrorism Database (GTD) for 1970-2020. TOQA enables large-N analyses of escalation, dormancy, and organizational change, covering 2,868 organizations while preserving inactive periods. It addresses three recurring data-construction problems: fragmented organizational identities, misleading inactive periods caused by miscoded lifespans, and weak integration between terrorism and conflict data. Across these challenges, TOQA systematically handles 984 GTD perpetrator labels through alias harmonization and structured exclusion of non-organizational entities. TOQA also links organizations to 439 UCDP Actor IDs, incorporating one-sided violence measures and facilitating integration with civil war and other conflict-related datasets. The resulting panel improves researchers’ ability to trace organizational trajectories over time, compare behavior across contexts, and draw stronger causal inferences by making sequencing, lagged relationships, and within-organization change observable. A Random Forest classifier illustrates one use of TOQA by forecasting whether an organization will record any fatalities in the next quarter. On the same matched sample, the integrated GTD-UCDP model outperforms a GTD-only model, illustrating the advantages of integration.

Intelligence organizations often fail to anticipate strategic developments even when relevant information is available because signals remain fragmented, ambiguous, overwhelming, or filtered through entrenched assumptions. The September 11 attacks, Russia’s 2022 invasion of Ukraine, and the October 7, 2023 Hamas attack illustrate how, even within some of the world’s most capable intelligence organizations, available indicators do not reliably translate into effective strategic warning, with outcomes shaped by differences in collection, interpretation, dissemination, and uptake. This article examines these failures and introduces AI-INT, or Artificial Intelligence Intelligence, as a response. AI-INT is an integrated, actor-specific synthesis-and-simulation layer within all-source analysis. It assembles and fuses relevant multi-source intelligence to build and update actor models, then uses them to run simulations at scale under systematically varied assumptions and conditions. It builds on wargaming’s value in deepening understanding, challenging assumptions, exposing blind spots, and exploring alternative trajectories by combining fused intelligence with LLM- and ML-based methods to expand the scenario space. As a bounded architectural illustration, the article presents SESBot, a Behavioral-AI framework for simulating the decision-making of selected Semi-State Terrorist Organizations (SESTOs). SESBot combines LLM-based agents with machine-learning behavioral anchors, illustrating one implementation of intelligence-grounded actor simulation within AI-INT. AI-INT’s promise lies not in predicting the future with certainty or replacing human judgment, but in helping analysts and decision-makers test assumptions, compare alternative scenarios, identify analytic blind spots, surface overlooked trajectories, connect them to monitorable indicators, and communicate uncertainty rigorously.