Palestinian residents survey the debris of structures demolished in the aftermath of an Israeli airstrike in Rafah, Gaza.
As the civilian death toll continues to rise in the conflict-ridden Gaza Strip, reports on Israel’s deployment of artificial intelligence (AI) in targeting Hamas militants are drawing increased scrutiny.
A recent article by Israeli media outlets +972 Magazine and Local Call highlighted the significant reliance of Israeli forces on AI tools such as “Lavender” and “Where’s Daddy” during the ongoing conflict.
While “Lavender” is tasked with identifying suspected Hamas and Palestinian Islamic Jihad (PIJ) militants and their residences, “Where’s Daddy” is designed to track these targets and notify Israeli forces upon their return home. According to the report, which references insights from six Israeli intelligence officers familiar with the AI systems, the approach includes targeting militants in residential areas rather than solely in military settings, often resulting in civilian casualties.
An unnamed officer revealed, “We were not interested in killing [Hamas] operatives only when they were in a military building or engaged in a military activity… the IDF bombed them in homes without hesitation, as a first option.” This strategy, facilitated by AI technology, has raised concerns about the significant risk posed to civilians, who are sometimes labeled as “collateral damage.”
The Israel Defense Forces (IDF) have emphasized their efforts to minimize harm to civilians within operational constraints, refuting allegations of intentionally targeting residential areas.
Challenges and Concerns
The utilization of the “Lavender” system has led to instances of misidentification, with approximately 10% of cases resulting in errors, as per insider accounts. These misidentifications often involve individuals sharing names or possessions with militants, contributing to the heightened risks faced by civilians.
Brianna Rosen, a senior fellow at Just Security, expressed concerns about the potential impact of Israel’s targeting criteria and AI inaccuracies on civilian safety, particularly in the fast-paced context of warfare.
Furthermore, reports suggest that human oversight in the target identification process is minimal, with officers reportedly spending minimal time verifying AI selections, primarily focusing on gender confirmation.
Ethical Implications
Critics have raised ethical concerns about the implications of automating target selection, highlighting the potential for biased and imprecise targeting practices that may contravene international humanitarian law.
The IDF’s preference for using unguided “dumb” bombs against lower-ranking Hamas members, as opposed to more precise munitions, has drawn criticism for its impact on civilian populations. President Joe Biden has cautioned Israel against indiscriminate bombing practices, underscoring the importance of upholding international norms.
International Law and Accountability
Questions have surfaced regarding Israel’s adherence to international legal standards, particularly the principle of proportionality, which aims to prevent disproportionate harm to civilians relative to military objectives. Observers have noted Israel’s broad interpretation of international law and the need for enhanced precautionary measures.
Despite the IDF’s rebuttals to allegations of indiscriminate targeting, concerns persist regarding the transparency and accountability of Israel’s AI-driven military operations.
Conclusion
While the revelations from +972 Magazine and Local Call shed light on Israel’s integration of AI technologies in combat operations, significant uncertainties and ethical dilemmas remain unresolved. The opaque nature of Israel’s targeting practices and the potential implications for civilian populations underscore the need for greater transparency, accountability, and adherence to international legal standards.
