Killer drones are being linked to AI in Gaza and Ukraine, fusing surveillance and death. Will a growing resistance set new rules for war?
The conflicts in Gaza and Ukraine have highlighted the pace and extent of the development and use of automated weapons systems. The extensive use of drones in Ukraine was an eye-opener, but the nature and scale of the use of artificial intelligence (AI) in automated weapons systems in Gaza, and the consequential appalling death toll, especially for women and children, has shocked the international community.
The first AI war
The eleven-day war in Gaza in May 2021 was Israel’s ‘first AI war’ using machine learning and advanced computing, according to the Jerusalem Post. The tools tested in the May 2021 war have been used extensively over the past six months. The Israeli-Palestinian publication +972 Magazine has reported on the use of an AI target-creation platform called Habsora, which in English means ‘the Gospel’ (one can only wonder who decided on this name and why, in the homeland of the Christian Gospel). It generates, among other things, automatic recommendations for attacking private residences where people suspected of being Hamas or Islamic Jihad operatives live. Israel then carries out large-scale assassination operations through the heavy shelling of these residential homes, resulting in an extremely high death rate for women and children.
According to the +972 article, Habsora processes enormous amounts of data from a wide variety of sources—including civilian sources—analyses it looking for any potential link to Hamas, and then provides potential targets for attack. According to former IDF Chief of Staff Aviv Kochavi, in the past ‘we would produce 50 targets in Gaza in a year. Now this machine produces 100 targets in a single day with 59% of them being attacked’. According to the IDF, in November, during the first 35 days of the war, Israel attacked 15,000 targets in Gaza. In comparison, in the 2014 war, which lasted 51 days, it struck between 5,000 and 6,000 targets, reported The Guardian.
Another platform called Lavender reportedly ‘analyzes information collected on most of the 2.3 million residents of the Gaza Strip through a system of mass surveillance, then assesses and ranks the likelihood that each particular person is active in the military wing of Hamas or PIJ’, with the machine giving ‘almost every single person in Gaza a rating from 1 to 100, expressing how likely it is that they are a militant’. Israel’s automated war on Gaza has cost the lives of over 34,000 Palestinians. These ‘killer robots’ truly are weapons of immense destruction.
It is only recently that the full horror of these new weapon systems has become visible. In April, four UN Special Rapporteurs deplored the purported use of AI and related military directives by Israel in occupied Gaza as leading to an unprecedented toll on the civilian population, housing, vital services and infrastructure. ‘We are especially concerned about the alleged use of AI to target “family homes” of suspected Hamas operatives, typically at night when they sleep, with unguided munitions known as “dumb” bombs, with little regard for civilians who may be in or around that home’, they said.
Nor are we uninvolved. Professor Richard Tanter from the University of Melbourne has written to the Australian government asking for an investigation into Australia’s possible complicity in genocide through the provision of intelligence sourced by the three Orion satellites that cover Gaza. The data from these intelligence satellites, which are controlled by Pine Gap, the US military base near Alice Springs, is sent via the United States to Israel.
The killer robot race is on
There are many ways in which AI systems can collect data about individuals: through loyalty cards, online purchasing, bank accounts and social media profiles, not to mention large-scale data breaches and the loss of personal data. The Gaza conflict has shown how this data can be utilised to provide targets for lethal automated weapons systems.
An International Committee of the Red Cross video drives home the issue: ‘If we don’t want our faces, our movements, our gestures, our communications, our friends, our location, or the number plate on our car to be the target for a bullet or a missile without any human judgement, governments must act now to ban autonomous weapons’.
AI is expected to transform all sectors of society, including the very character of war, according to the US Department of Defense (DOD): ‘Failure to adopt and effectively integrate AI technology could hinder national security’. As a result, the DOD is investing billions of dollars and making organisational changes, to integrate AI into its war-fighting plans.
Others making large-scale investments in AI systems include NATO, the United Kingdom, the United Arab Emirates, Saudi Arabia, Japan and South Korea. Little is known about investments made by Russia and China in autonomous weapons systems, but the United States believes that both are making significant investments in AI for national security purposes. The introduction of drones for surveillance and offensive purposes is occurring in conflicts around the world, including in Africa and even Myanmar, where drones are being assembled in remote bush camps by resistance forces.
Australia has joined the race, with a focus on the second pillar of the AUKUS agreement with the United States and United Kingdom. While it’s the nuclear-powered submarines that have grabbed the headlines, the second pillar of AUKUS focuses on ‘advanced capabilities’ and includes a range of high-tech applications: undersea robotics and autonomous systems; quantum technologies; AI and autonomy; advanced cyber technologies; hypersonic and counter-hypersonic capabilities; electronic warfare; defence innovation and information sharing. In May last year, it was revealed that Australia, the United States and the United Kingdom had held a trial of AUKUS advanced capabilities focused on AI and autonomy.
The international arms industry is keen to get a slice of Australia’s growing expenditure on weapons. The Australian Campaign to Stop Killer Robots has reported that the largest private manufacturers, including Lockheed Martin, DefendTex, Boeing, Thales and BAE Systems, are involved in the development and importing of military technology in Australia. Lockheed Martin has formed a close partnership with the Department of Defence and recently signed a $500 million contract to build Australia’s advanced integrated air and missile defence system. A US company, Anduril, has set up an Australian arm and has been recruiting technologists in military engineering. Anduril makes a software system called Lattice, an ‘autonomous sensemaking and command & control platform’, as well as producing flying drones, and has a deal to produce three robotic submarines for Australia. On 18 April 2024, Defence Industry Minister Pat Conroy unveiled the first Ghost Shark autonomous submarine prototype, saying it had come in early and on budget. Anduril’s Ghost Shark is part of the Australian military’s $5.2 to $7.2 billion investment in undersea uncrewed maritime systems.
The Israeli military technology company Elbit Systems has also set up in Australia and in 2024 won a five-year, $970 million contract from the federal government. One of the products Elbit makes is the Hermes drone, which is believed to have been responsible for the missile attack on the three World Central Kitchen vehicles in Gaza which killed an Australian and six of her colleagues.
The expansion of the global military industrial complex in Australia is having wide-reaching societal and economic implications, because much of its research and development capacity sits in civilian industries and universities. As defence expenditure has increased and university funding has declined, an interdependent relationship has transpired between higher education and defence, with the scales tipped in favour of ‘national security’. Through a growing number of federally backed defence research and ‘industry collaborations,’ universities have become deeply embedded in Australia’s extensive military buildup.
Calls for action grow louder
An open letter signed by 137 founders of robotics and AI companies from twenty-eight countries has urged the United Nations to urgently address the challenge of lethal autonomous weapons, often called killer robots. The open letter, coordinated by Professor Toby Walsh from UNSW, was the first time that AI and robotics companies have taken a joint stance on the issue.
There is growing international support for states to adopt new, legally binding rules on autonomous weapons systems. Based on statements made at the United Nations General Assembly and at the Convention on Certain Conventional Weapons, including meetings of the Group of Governmental Experts on emerging technologies in the area of Lethal Autonomous Weapons Systems, more than ninety countries have expressed support for legally binding rules.
Regrettably, however, the Australian government appears to be dragging its feet, suggesting that it ‘remains to be convinced new international law is required’. Stop Killer Robots campaigner Matilda Byrne has commented that while Australia has endorsed the use of possible measures in the design, development, deployment and use of autonomous weapons, it has insisted that these must not reflect a legal requirement.
The United Nations has convened a Summit for the Future in New York in September, and it is expected that Foreign Minister Penny Wong will attend. In preparation for the Summit, the Secretary General has released a New Agenda for Peace which covers a broad range of issues, including lethal autonomous weapons. The draft recommendation calls on member states to ‘conclude, by 2026, a legally binding instrument to prohibit lethal autonomous weapon systems that function without human control or oversight, and which cannot be used in compliance with international humanitarian law, and to regulate all other types of autonomous weapons systems’.
Australia must support this recommendation for two reasons. Firstly, it will take up a seat on the UN Peacebuilding Commission in 2025, and failure to support the prohibition of these weapons will undermine its international reputation as a nation committed to peace. More importantly, it is simply time to prohibit autonomous weapons for the sake of humanity.