Introduction
The acceleration of warfare to an industrial scale in Gaza is not a solo Israeli achievement. It also depends on a wider transnational ecosystem of military, technological, and industrial cooperation, including with European actors. While most European governments have issued formal statements expressing concern over the humanitarian catastrophe in Gaza and condemning the relentless bombardment and scale of military operations, many European companies have continued to supply components, technologies, and services that support Israel’s military capabilities, transitioning in the process from passive suppliers to active participants in the deployment of autonomous systems.
The European stance on Palestinian rights rings hollow against the backdrop of its own industrial supply chains. In 2024 and 2025, a historic wave of diplomatic recognition of the State of Palestine by nations such as Spain, Belgium, Norway, and France signaled a symbolic commitment to international law, bringing the total number of recognizing states to over 150. Yet, while these gestures suggest a major shift in Western policy, the flow of dual-use Artificial Intelligence (AI) and surveillance technology remains largely uninhibited. This political moment shields a systemic hypocrisy: diplomatic support for a Palestinian state exists alongside corporate contracts that arm the machinery of its dismantling. Even in Sweden, which recognized Palestine in 2014, celebrated diplomatic gestures coexist with a vast network of European corporate and institutional support that continues to furnish a key part of the technological backbone of the warfare that makes such recognition necessary.
For the past two years, the genocidal war on Gaza has moved into a paradigm of AI-driven destruction. Israel’s large-scale use of targeting systems such as Lavender and “Where's Daddy?” has turned family shelters and childhood homes into data points for execution. According to media reports citing Israeli intelligence officers, Lavender carries a known 10% error rate, meaning one in 10 people it flags is a civilian, a fact that did not deter its widespread use. Human oversight for these machine-generated lists was reduced to a mere 20 seconds per target, a window used solely to verify that the target was male before authorizing a strike. These tools are often programmed to wait for an individual to enter their residence, which means that strikes often occur at night, when multi-generational families are gathered together. This process of “automated domicide” has contributed to more than 71,000 Palestinians killed and 171,152 living with life-altering wounds. The human outcome is a catastrophic erasure of entire lineages in a single flash, compounded by the systematic destruction of the hospitals and schools that once held the fabric of Palestinian society together.
Corporate ledger sheets document a massive commercial benefit from this devastation. The Israeli company Elbit Systems, for example, reported a staggering revenue jump to nearly US$2 billion (around €1.7 billion) in late 2025, reaching an unprecedented order backlog of US$25.2 billion (around €21.7 billion). This is not an isolated success; by providing the sensors and data processing units that power these strikes, companies such as the German-based Hensoldt and French-Italian STMicroelectronics have turned a humanitarian collapse into a lucrative business opportunity. By converting documented human rights violations into a “combat-proven” brand and providing the Research and Development (R&D) backbone through programs such as Horizon Europe (which has granted over €1.1 billion – around US$1.3 billion – to Israeli entities), these European actors have become actively involved in a campaign of automated genocide.
This paper recenters the narrative on the essential European network that sustains this ecosystem to answer three main questions:
- How do Israeli defense companies use European subsidiaries to get around arms embargoes and tap into public funding?
- What is the actual scale of European Union (EU) research grants, such as Horizon Europe, in supporting the technology used in Gaza?
- How does the hosting of military targeting data and AI infrastructure on European servers implicate the EU as the digital architecture of the very practices its own regulations prohibit?
Methodology
The findings in this paper result from an investigative synthesis necessitated by the limited availability of public information. No comprehensive database exists to track the flow of European technology into the Israeli military and AI ecosystem. As a result, the researcher had to act as a forensic aggregator to overcome the shroud of secrecy maintained by both European states and private corporations. These entities frequently use national security and commercial confidentiality to hide their contracts from the public, effectively turning the digital supply chain into a black box.
To bridge this information gap, the researcher built upon the foundational work of organizations such as Amnesty International, Who Profits, AlgorithmWatch, and the United Nations Special Rapporteur on the Occupied Palestinian Territories. This paper consolidates fragmented pieces of evidence found in military trade journals, corporate press releases, and non-governmental organization field reports to create a unified registry of support to genocide.
For the purposes of this paper, corporate involvement is assessed through three analytical categories of contribution. Material contribution involves the supply of physical parts or components for weapons platforms. Financial contribution tracks the flow of R&D subsidies, public grants, or direct investment into military-linked firms. Enabling contribution identifies the provision of essential “dual-use” services, such as cloud infrastructure and the data processing capacity required for algorithmic targeting.
This paper's classification of material, financial, and enabling contribution follows the framework developed by the UN Guiding Principles on Business and Human Rights, which assess a company's connection to an adverse human rights impact through its own activities, products, or services, or through its business relationships. This is an analytical and human-rights-due-diligence framework, not a determination of criminal or civil legal liability, which only a court of competent jurisdiction can make.
The researcher adopts a framework rooted in Palestinian-centered epistemology rather than the techno-optimism that dominates mainstream Western and Israeli reporting. Where conventional coverage focuses on the sophistication and technical precision of AI systems, this paper shifts the lens toward accountability, treating the supply chains not as innovation but as infrastructure that renders Palestinian life machine-readable and targetable, erasing not only individual lives but Palestinians' collective existence.
This framework is designed to be replicable. The intersection of public subsidy records, corporate disclosures, and field-based human rights documentation produces a methodology that other researchers can apply to map the corporate and state structures enabling automated warfare across different geopolitical contexts. The template is scalable and testable, filling a persistent gap in conventional oversight, where these data streams are typically siloed.
Three primary groups of companies serve as the focus for mapping this ecosystem. One group consists of European-based corporations operating in the security, technology, AI, and military sectors, including their international subsidiaries. Another group includes Israeli companies in these same fields that receive funding from EU programs or maintain active research collaborations and joint ventures with European firms. The final group involves global technology companies that operate critical cloud infrastructure, data centers, or AI services within EU territory. A strict geographic focus remains on the EU and its member states.
The timeframe for analyzing operational data and corporate revenue surges spans from October 2023 to March 2026. A window of this length captures the transition from an economy of occupation to an economy of genocide, as defined by recent UN reports. It also allows for the tracking of sudden spikes in defense-related profits in correlation with specific military operations in Gaza. The scope for tracing R&D subsidies extends back to 2021. A longer window of this type accounts for the multi-year development cycles required for AI and defense technologies to move from a funded project to active deployment.
Specific inclusion criteria guided the selection of these companies. First, a firm must maintain a headquarters or infrastructure in the EU or participate in research collaborations with European partners. Second, a firm must have received EU-level R&D funding for security or AI projects at any point during the five-year development window. Third, the technology must be documentarily linked to the Israeli defense apparatus in Gaza through primary sources or investigative reports. Companies providing purely non-security-related infrastructure were excluded to keep the research focused exclusively on the technical and financial aspects most directly connected to military targeting and operational systems.
The research process followed a multi-layered verification protocol to establish high-confidence linkages. Baseline data collection began with an audit of corporate annual reports and investor disclosures to identify defense-sector revenue surges often omitted from public-facing marketing. Analysis of EU funding portals, specifically the European Defence Fund (EDF) and Horizon Europe, provided the second layer of verification regarding public grants for surveillance and AI development. The evidentiary threshold required successful triangulation between these financial records and the operational reality documented in human rights reports or investigative journalism.
Indirect involvement was assessed through the presence of critical components or digital infrastructure essential to the functioning of larger autonomous systems. Inferences regarding military application were made by matching specific technical capabilities, such as high-frequency sensor resolutions or cloud processing speeds, with the known requirements of Israeli defense operations in Gaza. Any entity lacking a direct documentary link or a clear technical necessity for military use was not included in the registry.
The resulting registry in Annex I classifies each entity according to the three categories of contribution defined above. Each company entry identifies whether its involvement constitutes material contribution, financial contribution, enabling contribution, or a combination of these categories. This classification is designed to serve as an actionable tool. Litigators pursuing corporate accountability can isolate entities by contribution type. Parliamentary committees reviewing export licenses can identify which category of support persists despite declared embargoes. Divestment campaigns can prioritize targets based on the directness of a company’s contribution to the kill chain. The methodology also provides a specialized architecture for tracking the migration of civilian AI into systems used for surveillance, mass destruction, and targeted killing. Future investigations can adapt this specific layering of subsidy records and operational data to audit other high-risk supply chains where corporate accountability remains hidden.
The views represented in this paper are those of the author(s) and do not necessarily reflect the views of the Arab Reform Initiative, its staff, or its board.
