Why AI will not replace human strategic judgement in war
Is AI about to automate war? Will autonomous drone swarms and killer robots controlled by AI dominate the battlespace and determine the winner? In AI, Automation, and War, Anthony King debunks this science fiction–tinged narrative of AI’s military potential, exploring instead the actual applications of AI by the armed forces over the last decade. He finds that AI is not going to replace human commanders and combatants; the machines are not about to take over. Rather, the military has used, and will continue to use, AI to process data at a scale and speed that exceeds the capacity of humans. AI will be used primarily to improve military understanding and intelligence.
King explains that military commanders, enabled by the data processing power of AI, will be able to see the battlespace at a previously unattainable depth, fidelity, and speed. AI will help the armed forces plan, target, and conduct cyber operations faster and more effectively. In order to harness AI in this way, however, a radical organisational transformation is taking place. The armed forces are integrating civilian technologists into operational headquarters to work alongside military staff. This partnership between the armed forces and the technology sector signals the emergence of a military-tech complex that promises to be as powerful in this century as the military-industrial complex was in the last.
King proves that even when you agree with an argument, the author can make the reading of it unbearable. With a prose that is as tedious as it is rambling, and examples that indicate (at best) a cursory level of research, the greatest contribution the author makes to the space is his main idea: AI will not so much be transformational, as evolutionary, to modern warfare. Indeed, counter-intuitively, AI enabled warfare might actually slow down - not speed up - fighting between states in the future. The idea saves the book from totally failure, but this would have been much better done as a 20-page article in an academic journal instead of a 183 page book.
Artificial Intelligence or AI is a real technology buzz word. It has dominated global news stories and is seen as the great new leap forward in modern technology that will have a transformative effect on all our daily lives. Whereas the advantages to the civilian sector seem positive, good and acceptable. There has been a darker cloud over the military use of AI in terms of its morality. Also, most worries about the dangers of AI getting out of control involve military use. Will we be taken over by an out of control military controlled automated system that has the capability of destroying humanity?
The science-fiction ideas of out of control drone swarms and automated human-killing robots are on the whole debunked by author, Anthony King, in this book that focuses mainly on the actual use of artificial intelligence in war and how this has developed over the past decade or so. He stresses that humans play and will in his view, always have to play, an essential, integral part of any AI on the battlefield. For future conflict there will be a growing partnership between militaries and the civilian tech sector with Big Tech playing a prominent role in warfare and inevitably being drawn into decision-making and polticising the theater. We have seen how Elon Musk became involved in the Russo-Ukrainian conflict by providing the Ukrainians with free Starlink capabiilities which powered up their tech capacity with his state-of-the-art internet satellite system, thus enabling their defences against Russian aggression to succeed. Musk became politically active and, fearing an escalation into wider nuclear conflict by what he saw as a potentially unethical application of this mainly donated technology, flicked the switch and powered it off to prevent the Ukrainians from attacking Russian Navy fleets in Crimea. Google, Amazon Web Services and Microsoft are all qunitessential global tech leaders that now have prominent roles in modern warfare. The civilian tech operators who design, run and programme AI – the essentia`l software engineers, are now being located in forward military bases towards war-fronts and are fusing into command and contropl set ups. In essence AI is now a partner in the millitary Chain of Command. It may not hold independent agency and is, in essence, just a tool. But the reliance on it, and its amazing capability to cut out time spent on difficult planning tasks, makes it core technology to military planners, decision-makers and commanders. The main use of AI is in its ability to quickly process large data sets. Vast data from drone surveillance, CCTV footage, civilian reports and documentation of enemy troop movements, terrain maps, gathered enemy intelligence, are all fed into the AI machine and results are churned out at the ‘Speed of Light’. Obviously, AI is still very much in its infancy and there are potentially huge problems with hallucinations or errors. What is fed into AI ultimately controls what is fed out. Military chains of command are having to learn how to use the software effectively. How to program it, how to harness it, what its capabilities are and how to advise the active engineers on refining it in real-time often, in order to maximise its potential and effectiveness in modern warfare.
The Israeli military, IDF (Israeli Defence Forces), are among the most advanced military forces in terms of their AI capability. With national consciption in Israel for all citizens, the IDF have a natural civilian / military partnership, built into their military actions in war. Israel has monitored the West Bank and Gaza for years, focussing on terrorist activity and potential enemies of their State. Unit 8200 of the IDF is a specialist cyberwarfare unit that employs AI. They process large data sets of surveillance of Palestinians and identify terrorist cells and potential threats and if necessary take them out using military action. We have seen the IDF employ operations such as Operation Protective Edge, Gospel, Guardian of the Walls, Swords of Iron and Lavender. Unfortunately, as with any new military tech, the enemy are always strategising against its use and focus on exploiting weaknesses and flaws in any new system. At the outset of the current ongoing war between the IDF of Israel and the terrorists Hamas of Palestine, prior to the Hamas attack on Israeli citizens on 23rd October 2023 where they massacred, raped and mutilated 1200 Israeli citizens and took 200 hostages, Hamas had conducted a drone operation of their own to take out and offset the tech capability of the IDF defences. With a much lower-tech limited drone capability they attacked surveillance cameras and defence posts in the vicinity of their target zone and this caused a blur in the AI defence system so that when the Hamas terrorists went in, it was done as ghosts and they had exploited a gap in defences. They could get in and get out and the IDF’s tech was only able to piece together and analyse what had happened during the event after it had occurred due to the hole that Hamas had created in their defences.
Closer to home in the United Kingdom, our own military forces are also embracing tech and AI. An example of the British Army’s use of AI was during the COVID-23 epidemic when the army were brought in to assist in management of the crisis in the City of Liverpool. in 2020. 8 Engineer Brigade under Brigadier Fossey were drafted in to assist Liverpool’s Covid Gold Command in mass testing and treating the city’s 1.6 million population and to help organise the epidemice spike in this poor urabn area. AI was used by the Army in their civilian-military partnership in order to straetgise and plan and implement testing sites and to quickly cover the entire population and to enable the COVID outbreak to be successfully controlled and managed. Although, strictly speaking this was not a war operation. The skills used in the operation and the use of AI was very much akin to how it would be used in a battlefield situation whereby enemy could be targetted and destroyed. Militaries use AI in ordewr to strategise and plan operations. It has become an indispensable tool in preparing and analysing how to proceed quickly in any given operation and deeply now affects nearly all central command and control operations as well as those actually out in the field. We are seeing the necessity of military commanders and soldiers to all gain new skills in tech and civilian partnerships both in the field and remote from the battlefield itself are the new norm.
In the US, AI has been embedded in the military. Renowned projects are Project Maven and a lot of AI work in the military is overseen by the JAIC or Joint Artificial Intelligence Center. US military has become reliant in the theater on military-focussed tech companies such as Palantir which have successfully capitalised on their business model and are a key element now of many US operations and are at the cutting cusp of modernising drives in military technology, focussed on the successful harnessing af AI and its practical implications.
In this book, the author examines in detail the actual use of AI, its advantages and its failures. He never gets too dreamy or reaches into the realms of science-fiction and scraemongering. If anything, there is a focus on the limitations and potential failures of AI. He is in that sense quite skeptical but equally very thought-provoking. The information presented in this book does make you rethink modern warfare and indeed future warfare. I recently studied at a Tel Aviv Israeli based Cybersecurity education Institute (Masterschool). Out of that work, sprang my own #cybersecurity / #cyberwarfare business venture : Four Four Cyber https://x.com/fourfourcyber #44C I have since the end of my study at masterschool started to explore more AI and have contacted various militaries and tech companies such as Palantir in order to seek business investment and to learn and strategise more in my own work and to develop decent technology that could be used in war if necessary. This book has been part of my education for Four Four Cyber and I would seriously recommend it to anyone with an interest in Artificial Intelligence, in particular in the military application of AI.
The book offers an empirical description of AI, which is then used to argue that AI will be a service for the armed forces to enhance key capabilities in warfare. The author also argues that AI will not become lethal itself, and he argues that a new habitat is emerging, where armed forces integrate more geeks and nerds into their ranks to be able to use the service.
It is easy written, but it repeats some arguments too much, while some other arguments are not fully developed. Yet, with its empirical examples a fascinating book, offering far more insights than only on war. The arguments of the book can be used also considering the debate on future unemployment due to AI. Definitely a recommendation.
This entire review has been hidden because of spoilers.