Israel's military campaign in Gaza has seen an "exponential" increase in the use of artificial intelligence, with an AI-powered analysis tool known as Lavender reportedly generating a list of 37,000 potential targets for air strikes in the first weeks of the war. This deployment, detailed by progressive Israeli outlet +972 Magazine, marks a stark escalation in the integration of autonomous systems into lethal decision-making, raising immediate concerns about civilian harm. The International Institute for Strategic Studies (IISS) analysis suggests that while such systems allow for quicker decisions, they also increase the risk for errors, as targets are identified faster than humans can possibly assess them. Israel has rejected suggestions it was using an AI tool "that identifies terrorist operatives or tries to predict whether a person is a terrorist," stating analysts must conduct "independent examinations."
The Algorithm of War
Australia's defence force is actively pursuing similar integration, with Defence Minister Richard Marles recently meeting representatives from tech giants like OpenAI, Anthropic, Microsoft, and Google in the United States. The goal is to integrate AI into military operations, from intelligence assessments to piloting platforms with lethal payloads. One year ago, over the remote desert of South Australia, the Australian-made Ghost Bat aircraft, an autonomous system, successfully shot down a target with an air-to-air missile. There was no human pilot in the cockpit, nor a drone operator on the ground. Air Marshal Stephen Chappell, chief of Air Force, described the Ghost Bat as a "pathfinder for the integration of autonomous systems and artificial intelligence to create smart human-machine teams." He stated that while AI was involved in its design, the aircraft operates off "deterministic programming," choosing from thousands of pre-designed options. An operator in an E7 Wedgetail aircraft gave the final authorization before the missile launch in the test-firing, Chappell confirmed.
Capital's New Frontier
Beyond visible platforms like the Ghost Bat and its underwater counterpart, the Ghost Shark, AI's deeper influence lies in informing critical decision-making and intelligence analysis. Aina Turillazzi from the Australian National University’s Strategic and Defence Studies Centre noted that "most of what AI does in the military is upstream of any weapon," assembling situational pictures for human decision-makers. The Australian Defence Force (ADF) has adopted technology such as Palantir’s Maven Smart system, a move that has generated contention over the political links of its founders and concerns about data sovereignty. Malcolm Davis, a senior analyst at the Australian Strategic Policy Institute, described Maven Smart's role in conflicts, including the US-Iran war, as that of a "silicon commander," rapidly assessing battle spaces and reducing staff needs. The ADF previously told parliament that Maven Smart was being used in a "sandbox" environment, not yet connected to broader defence networks.
The State's "Guardrails"
Despite the rapid push for AI integration, concerns about human involvement persist. One month ago, the United Nations and International Committee of the Red Cross (ICRC) issued a joint call for a ban on "unpredictable" autonomous weapons that target human beings, raising fears that "life-and-death decisions will be taken by machines." ICRC legal advisor Georgia Hinds observed that promises of reduced civilian casualties with advanced warfare often prove false, with conflicts continuing to cause "unacceptable levels of civilian harm." Defence’s existing AI policies state that a human operator will always be responsible and accountable for any action carried out by artificial intelligence, ensuring someone maintains oversight and intervenes at critical points. However, Turillazzi warned of automation bias, where decision-makers under pressure might prioritize AI recommendations over their own judgment, especially in fast-moving conflict environments. She referenced Stanislav Petrov, a Soviet lieutenant colonel who, 43 years ago, averted a potential nuclear disaster by questioning automated warnings. Defence Minister Marles, when asked about AI adoption by potential adversaries, stated, "We need to be advancing our own technology so that friends and partners and allies have the technological advantage here," framing the arms race as a path to "making the world a safer place."