Israel’s recent military campaign in Gaza has been characterized as the first major “AI war,” utilizing advanced systems to generate lists of targets, purportedly linked to Hamas and Islamic Jihad. These AI systems processed billions of data points, assessing the likelihood that individuals were combatants. This reliance on **artificial intelligence** marks a significant shift in military strategy, reminiscent of the “fog procedure” established during the second intifada, where soldiers would fire into darkness under low visibility, justifying such actions as deterrence against unseen threats.
The parallels between the fog of war and the opacity of algorithmic targeting are striking. Both reflect a chosen blindness—an operational decision that obscures accountability and shifts the responsibility for violent outcomes from individuals to procedures. In this context, the AI systems develop targets by assigning probability scores, often with outdated information. One example is the tragic strike on the Shajareh Tayyebeh elementary school in Iran, where at least 168 people, primarily children, were killed. This incident highlights a critical flaw: the **intelligence** used to justify the attack was outdated and failed to reflect the school’s transition to civilian use nearly a decade prior.
The targeting systems in question did not merely malfunction; they operated under systemic conditions designed to prioritize speed and efficiency over accuracy. In the initial phase of the U.S.-Israeli strike campaign in Iran, AI systems reportedly generated thousands of targets in a matter of days. This allows military operations to occur at a tempo unattainable by human analysts alone, raising significant ethical and legal concerns. The speed at which these decisions are made—often in seconds—diminishes the role of human judgment, reducing operators to mere rubber-stampers of machine output. In the case of the Minab school, the process lacked mechanisms for flagging outdated intelligence, leading to catastrophic consequences.
The ramifications of such AI-driven warfare extend beyond individual strikes. Data reviewed by multiple outlets, including the **Guardian**, reveals that of over 53,000 recorded deaths in Gaza, only about 17% were identified as militants. This statistic underscores a troubling trend: a high civilian casualty rate associated with military operations often framed as precise. When AI systems automate targeting decisions, they inherit and encode existing biases into their frameworks, perpetuating a cycle of violence that prioritizes efficiency over human lives.
As AI continues to shape military strategy, questions of accountability become increasingly complex. The intertwined roles of private defense contractors and military operations blur the lines of responsibility. Companies like **Palantir**, integral in providing AI infrastructure, operate with a level of impunity, their systems embedded in military decision-making processes without adequate oversight. This exemption raises ethical questions about the companies’ roles in conflict, especially as their technologies integrate more deeply with lethal targeting systems.
Who is Responsible?
The accountability framework governing military actions has been rendered structurally irrelevant in this new landscape. Legal obligations to verify targets and ensure compliance with international humanitarian law are undermined when machines are allowed to dictate lethal force. The **EU AI Act**, while ambitious in its goals, conveniently exempts military applications, leaving gaps in regulation that could otherwise constrain the use of AI in warfare. As governments push for faster and more lethal capabilities, the potential for unregulated AI warfare grows.
The military-industrial complex’s integration with AI technology necessitates a re-evaluation of existing legal frameworks. The recent events in Gaza and Iran serve as cautionary tales, highlighting the urgent need for regulation that holds accountable both military and private entities involved in warfare. As nations invest in AI-integrated military capabilities, the imperative for oversight becomes increasingly critical. A future without a robust regulatory framework risks repeating tragedies like those seen in Minab, where outdated intelligence and rapid-fire decision-making led to devastating loss of life.
In this era of algorithmic warfare, the need for transparency and accountability is more pressing than ever. AI targeting systems must not only yield accurate outcomes but also do so through processes that can be audited and understood. The challenge lies in ensuring that the evolution of military technology does not outpace our ability to regulate and hold accountable those who wield it, both in the **Pentagon** and the private sector. Without decisive action, the fog of war may only deepen, obscuring the line between combatants and civilians and rendering accountability a relic of the past.
See also
Nvidia’s Price Exceeds DCF Estimate by 35.5% Amid AI Chip Demand and Policy Risks
Plugable Launches TBT5-AI Enclosure for Local AI Workloads at $599.95
AI Study Reveals Collaboration Boosts Human Creativity in Design Tasks by 800 Participants
