AI and Warfare
What’s this about: The Russian invasion of Ukraine has propelled war onto our screens and front page once again, making it an unwelcome but current topic of conversation and undesirable attention.
But how is AI affecting war?
Undoubtedly, war is changing. Throughout thousands of years, humans have gone from relying on spears, to gunpowder, to nuclear weapons. The next big step in warfare? Artificial intelligence. AI is impacting every aspect of military operations and technology. But how? Let’s take a deeper look.
Go Deeper to Learn More -
AI Arms Race
Weapons technology has evolved dramatically over the last 20 years. With the incredibly fast development of AI technologies, we are seeing it impact nearly every part of the battlefield, including logistics, weapon development, battlefield analysis, and training simulators.
The world’s most militarily-equipped nations, such as the United States, China, and Russia are involved in a new AI arms race. The U.S. Department of Defense has created the Joint Artificial Intelligence Center (JAIC) to scale AI and harness its power, while Xi Jinping has outlined that the People’s Liberation Army (PLA) must “accelerate the development of military intelligentization, [and] improve joint operations capabilities and all-domain combat capabilities based on network information systems” to fulfill China’s military development goals.
Russia, which is embroiled in a world conflict following its invasion of Ukraine, should not be underestimated. While the nation has been playing catch-up, falling behind the U.S. and China in AI research and development, it has been consistently working to create AI for electronic warfare, surveillance, intelligence, strategic decision-making, and more.
With new conflicts arising across the globe and an intense AI-arms race, it is more important than ever for the public to understand just how AI technologies are being implemented into the military. This is especially true given the public’s sentiment towards the use of AI in war. As one can expect, more people would rather AI fight wars. (Just check out the World Economic Forum poll below.)
AI in Weapon Development
Let’s start by looking at how AI has already made it onto the battlefield. None of the autonomous weapons systems have been more consequential and subject to controversy than autonomous drones. While drones have been a significant part of warfare for years, humans often remotely controlled them. But AI has enabled image-recognition and autopilot software, making them easier to mass-produce.
These autonomous drones were deployed in the Syrian and Libyan civil wars, and Turkey has used them to patrol its border. In March 2021, the United States Security Council published a report on the Second Libyan War that described the first known case of an AI-powered autonomous weapon killing people on the battlefield.
One of the most well-known examples of such a weapon is the Kargu-2 system, an unmanned aerial vehicle that sits above potential targets. It can detect radar signals from enemy systems before descending into attack mode, and the system can be operated both manually and autonomously using “real-time image processing capabilities and machine learning algorithms.”
The Kargu-2’s AI technology is designed and produced by the Turkish defense contractor STM. The drone relies on STM’s unique flight control system, and it consists of an Attack Drone Platform and a Mobile Ground Control Station.
These types of autonomous drones have also made an appearance in the current Ukraine-Russia conflict. The Ukrainian Armed Forces have recently announced that they launched the TB2 Bayraktars, another Turkish-made, tactical unmanned aerial vehicle. The drone is specialized in carrying anti-tank missiles, and it has been deployed in several other conflicts over the years.
Military Logistics
Militaries around the world are starting to use AI for logistics as well. The U.S. has been slow in this regard, opting to focus on other areas of the military. With that said, they have taken several steps forward since the creation of the JAIC.
For example, the U.S. Army has contracted IBM to provide cloud services, software development, and cognitive computing. Through its cloud services, IBM enables the Army to predict vehicle maintenance failures from more than 5 billion data points of on-board sensors. The Army has also adopted Watson IoT services to analyze unstructured, structured, and sensor data directly from military assets. All of this is made possible by AI-driven data analysis, which involves the use of machine learning algorithms to monitor and analyze huge amounts of data.
(To learn more about machine learning techniques and the different types of data, check out our article “DL Series: What is Deep Learning?”)
The U.S. Army has also contracted the development of an AI system that could monitor, diagnose, and treat battlefield victims with traumatic injuries. The Automated Ruggedized Combat Casualty Care (ARC3) system, which is currently being developed by intelligent systems solutions provider Charles River Analytics Inc., is a software that helps medics diagnose injuries, provide treatment when evacuation is not possible, and monitor patients.
Battlefield Analysis
Besides weapon development and military logistics, AI is also being used to improve battlefield analysis. This is crucial given the importance of fast decision-making during conflict, which can drastically change the outcome.
One example of this battlefield analysis is the Tactical Intelligence Targeting Access Node, or TITAN. Developed by Raytheon Technologies, it is an AI-enabled effort that helps operators detect, identify, process and engage targets faster than ever before. It also enables them to detect these targets at longer distances. TITAN consists of frontline forces and headquarters commander, the latter of which is a tactical ground station that can analyze massive amounts of incoming sensor data to detect threats.
In China, the PLA is developing deep learning systems that can analyze the graphical terrain and enemies’ posture, which will enable them to determine the safest route to maneuver to the frontline or behind defenses. The country is also exploring how deep learning can be used on other fronts, such as for the analysis of satellite imagery and other sensors to improve early warning capabilities.
Russia might be the country most focused on battlefield analysis, with the nation heavily investing in preparing for “information wars.” Researchers at the U.S. government-funded Center for Naval Analyses have suggested that Russia could use A.I. in its current conflict with Ukraine to analyze battlefield data, such as surveillance footage from drones.
Ukraine has also set out to use AI in battlefield analysis, receiving free access to the controversial facial recognition technology and database developed by Clearview AI. The nation is relying on it to uncover Russian assailants, identify the dead, and combat disinformation.
(Speaking of disinformation and AI, did you see the deepfake of Ukrainian President Volodymyr Zelensky calling on the Ukrainian forces to lay down their arms?)
(For more on this highly disruptive technology, read our piece on AI and deepfakes here.)
Military Training
AI technologies are also being used outside the battlefield, long before any conflict. You probably have an idea of how the military trains for warfare, but AI is drastically changing what that looks like. Military AI training software is playing an increasing role due to its incredible ability to mimic real-life combat situations while minimizing injuries.
Since long before it was mainstream, the U.S. military relied on virtual reality (VR) technology for training, simulation, and even therapy. VRs work best at simulating real-life warfare without the risks, and they help armed forces collaborate, handle mock weapons and improve their decision-making processes in highly stressful situations.
Let’s take a look at some of the specific ways VR is used in military training:
Situational awareness: VR training helps military personnel learn the skills needed to operate in high pressure and dangerous environments. They can improve combat and decision-making skills by reacting to various simulated scenarios.
Unit training: With VR, units can learn to work together before ever stepping on a battlefield. By the time they arrive, they will have a better understanding of their actions and how they can affect those next to them.
Activity-based training: By combining VR with haptic technology, the military can use it for activity-based training like weapon handling and parachute jumps. Personnel can also become better accustomed to working in tight environments, such as tanks or submarines.
VR and artificial intelligence go hand and hand. VR relies on AI and machine learning techniques to generate credible behavior for non-human roles and develop behavioral models. These techniques are also responsible for bringing to life the various simulation environments that soldiers encounter. Technologies like these will continue to evolve to prepare the military for modern conflicts, which will look a lot different than what we’re used to. More on this fascinating topic soon…
Becoming the Ruler of the World
Russian president Vladimir Putin once said whichever country becomes the leader in artificial intelligence “will become the ruler of the world.” That could very well be true, especially when it comes to AI and military power.
The AI arms race is propelling many of the world’s top militaries to pour funding into all sorts of areas like weapon development, logistics, battlefield analysis, and training. Similar to nuclear weapons, AI is quickly becoming a new way to measure power between superpowers.
At the same time, modern warfare is becoming far more complex, involving remote warfare, cyber attacks, deep fakes, and much more. The future of warfare is AI, not boots on the ground. And whichever nation most effectively implements it into their military will have the upper hand for years to come.