TLDR: The Israeli military has reportedly deployed advanced AI systems, including ‘Lavender,’ ‘Gospel,’ and ‘Where’s Daddy,’ in its operations within the Israeli-Palestinian conflict. These systems are used for identifying and targeting suspected militants, leading to significant ethical and legal concerns regarding their accuracy, potential for bias, and the resulting civilian casualties. Critics are calling for greater transparency, robust ethical frameworks, and human oversight in the application of AI in conflict zones.
Recent reports and investigations have brought to light the controversial and expanding role of artificial intelligence technology in humanitarian crisis management, specifically within the context of the Israeli-Palestinian conflict. The Israeli military has reportedly integrated sophisticated AI systems, such as ‘Lavender,’ ‘Gospel,’ and ‘Where’s Daddy,’ into its operational strategies. These AI tools are designed to sift through vast amounts of intelligence, including intercepted communications and surveillance data, to identify suspicious patterns, predict threats, and ultimately select targets for military action.
An Associated Press investigation, published in February 2025, revealed that the use of commercial AI models from major U.S. tech companies like Microsoft and OpenAI by the Israeli military significantly escalated following the October 7, 2023, attack. This investigation uncovered details of Microsoft’s confidential $133 million contract with the Israeli military and noted a nearly 200-fold increase in the use of commercial AI technology. Concerns have been raised that these tools, not originally developed for warfare, are now directly influencing decisions of life and death, leading to a surge in civilian casualties. The AP’s reporting specifically linked AI-driven targeting to the wrongful killing of civilians, including a Lebanese family with children.
Human Rights Watch, in a September 2024 report, highlighted that four digital tools used by the Israeli military in Gaza rely on faulty data and inexact approximations, potentially leading to violations of international humanitarian law, particularly concerning the distinction between military targets and civilians. These tools reportedly use Palestinians’ personal data, collected prior to the current hostilities, to inform military actions and identify targets. The reliance on machine learning, which draws inferences from data without explicit instructions, raises questions about the accuracy and potential biases embedded within these systems. Critics argue that instead of minimizing civilian harm, these digital tools may be exacerbating the risk to non-combatants.
Further revelations from Israeli media, reported in April 2024, detailed the ‘Lavender’ AI program, which reportedly identified 37,000 potential targets. This system’s decision-making process, with a reported acceptance of a 90% accuracy rate and a 10% margin of error for targeting, has been described as dehumanizing, particularly for Palestinian men who are broadly considered legitimate targets. The ‘Where’s Daddy’ system is reportedly used to track targeted individuals to their homes, further raising concerns about the precision and ethical implications of such targeting.
Also Read:
- Department of Defense Integrates AI into Operations Amidst Broader National Strategy
- Navigating the AI Frontier: Legal Ethics, Risks, and Policy in a Transforming Profession
Experts and human rights organizations emphasize the urgent need for transparent ethical frameworks, robust accountability mechanisms, and significant human oversight in the deployment of AI in conflict zones. There are growing calls for international pressure on the UN and its member states to regulate AI and cloud technologies as dual-use technologies, given their potential for both civilian and military applications, and to ensure that tech companies are held accountable for their role in enabling human rights violations.
