The impact of artificial intelligence on military defense and security

1. Emerging disruptive technologies represented by artificial intelligence are ushering in a new era

##From artificial intelligence, robotics to distributed Distributed Ledger Technologies (DLT), Internet of Things (IoT), emerging disruptive technologies have opened up a new era of commercial innovation. The social and economic changes caused by technological changes have a huge impact on the development of Canada's military power. These technologies include artificial intelligence and machine learning, quantum technology, data security and computer-aided hardware.

As a force multiplier, artificial intelligence can reshape the rules of war. In the context of great power competition and a multi-polar world, artificial intelligence is becoming the focus of competition. As guidance from the North Atlantic Treaty Organization (hereinafter referred to as "NATO") states, "Artificial intelligence technology is critical to the military defense and security of Canada and its allies." Currently, data and data-driven technologies occupy the commanding heights of the global economy. Competition in the global data economy is inseparable from competition among major powers.

China, Russia, the United States and many other countries are actively exploring artificial intelligence and its applications, with a focus on national defense and national security. At present, NATO is still the leader in the field of artificial intelligence and has strong technological advantages, and China is quickly catching up. The Chinese government hopes to lead the world in artificial intelligence by 2030 and expand its leadership in the industrialization of artificial intelligence by fully utilizing large amounts of data. Although the United States has established a leading position in artificial intelligence, China is expected to dominate the industrialization of artificial intelligence in the future. Because China not only has advanced commercial capabilities, but also has a sustained national strategy.

2. Artificial intelligence is a kind of dynamic innovation

People have discussed a lot about the concept of artificial intelligence. To be precise, artificial intelligence is not only a dynamic innovation , specific technologies and innovations, and a combination of advanced technologies. Currently, artificial intelligence technology has become the basis for many important applications, including web search, medical diagnosis, algorithmic trading, factory automation, ride-sharing and autonomous driving.

Artificial intelligence research began in the 1940s. With the improvement of machine learning and computer processing power, great interest in it has been aroused. The development of artificial intelligence is similar to the multi-level learning and reasoning capabilities of the human brain. When combined with big data and cloud computing, AI can “know” digital technologies by connecting “intelligent” systems and devices to 5G networks.

As a subset of artificial intelligence, machine learning is the most typical application of artificial intelligence. Machine learning uses statistical techniques to enable machines to "learn" without explicit instructions, drive applications and services, and increase the automation of analysis. This process of automatically improving performance through data is called "training the model." The most common form of machine learning is deep learning, which uses multi-layered artificial neural networks to replicate artificial intelligence. Deep learning architectures such as deep neural networks, recurrent neural networks, and convolutional neural networks can support computers in a wide range of research areas such as vision, speech recognition, machine translation, and natural language processing.

3. The United States and Canada invest heavily in building artificial intelligence

Artificial intelligence is the core of emerging disruptive technologies. Currently, the United States remains the global leader in artificial intelligence. The National Science Foundation (NSF) invests more than $100 million in artificial intelligence research every year, and the Defense Advanced Research Projects Agency (DARPA) recently proposed the "Next Generation Artificial Intelligence" project , plans to invest US$2 billion with the goal of improving situational reasoning and adaptive reasoning capabilities.

Canada has also been a leader in the field of artificial intelligence. Under the guidance of the 2017 "Artificial Intelligence Strategy", Canada's artificial intelligence ecosystem has developed rapidly, and the government will increase defense spending every year and focus on Focus on emerging disruptive technologies. At present, the Canadian government has promised to invest a lot of money in the research and development of artificial intelligence. In the past 10 years, the Canadian government has invested US$443.8 million. According to the Canadian government’s 2021 budget report, US$185 million will be used to support the commercialization of artificial intelligence research; US$162.2 million will be used to recruit top academic talents nationwide; US$48 million will be used for research institutes; US$40 million will be used for Enhance computing capabilities for researchers at the National Institutes of Artificial Intelligence in Edmonton, Toronto and Montreal; $08.6 million to promote the development and adoption of standards related to artificial intelligence.

Unlike the development of traditional military technology, no country can monopolize the military application of artificial intelligence. Extensive collaboration between researchers and industry means that artificial intelligence and machine learning will be further adopted globally, so it is likely that many future military applications will directly adopt technologies developed for commercial use.

4. Military applications of artificial intelligence

Artificial intelligence is a field that can have a wide impact on commercial and military technology fields. The widespread use of artificial intelligence means that the technology can re-align the pace and organization of modern forces. Taken as a whole, AI represents a tectonic shift in the nature of national security. Therefore, future military applications will focus on research and development, acquisition, and integration of advanced and transformative technologies, including cyber and autonomous systems.

4.1 Improving intelligence capabilities

Artificial intelligence, which appears as a war tool, can ensure Canada’s national security, especially improving its intelligence capabilities.

Warfare in the digital age is increasingly becoming knowledge-based. When conflict enters the information domain, military planning will focus on information/disinformation operations, cyber operations, intelligence operations, and political or economic influence operations. In fact, hybrid warfare has long been used as a war tool. Its purpose is to use network propaganda, destruction, deception and other non-military operations to weaken the enemy from within.

Networks have always been a major target of attacks by adversaries, states, criminal organizations and non-state actors involving surveillance and reconnaissance, intelligence and sensitive information. The development of technology has greatly broadened the scope of access to data and information. Currently, most of the information that drives strategic intelligence is Open Source Intelligence (OSINT) or public resources.

Modern warfare is critically dependent on secure, timely and accurate information. As information grows exponentially, data analysis becomes increasingly difficult, prompting the adoption of new analytical models and web tools. In the digital age, intelligence personnel urgently need new platforms, new tools and cross-domain OSINT, and artificial intelligence can meet this demand. Artificial intelligence and machine learning can greatly improve Canada's national intelligence capabilities by sifting through large amounts of data. Although artificial intelligence systems cannot provide causal analysis, they can greatly improve intelligence in data management and data-driven analysis.

4.2 Reshape the military command and control system

Artificial intelligence has changed the old military conflict model. Facing the data-driven battlefield, decision-makers should promptly adjust their security posture. Currently, the Canadian Department of Defense An important challenge facing the military and military is that data-driven networks are reshaping military command and control systems at an extremely rapid rate.

The advantage of an integrated system is that it can efficiently coordinate military operations. In the military command and control system, people and sensors perform threat detection and push information to the decision-making stack so that decision-makers can respond accurately, but integrated The command and control system also means that a single point of failure will become a weak link and be attacked. "Top-down" decision-making is difficult to adapt to emergency challenges in complex situations, and the application of artificial intelligence will further accelerate the decision-making process. Therefore, It poses a challenge to the traditional military command and control system.

Innovations in neural computing, generative adversarial networks, artificial intelligence decision support, data and intelligence analysis will have a huge impact on military operations. In the digital age where platforms, technologies and applications are integrated, artificial intelligence and machine learning are critical to consolidating and strengthening military power. Artificial intelligence is not a single technology, but consists of a series of technologies that can be integrated into a variety of military and commercial applications, and data is the basis for the continuous development of these technologies. Digital technology is driven by data and further drives the development of artificial intelligence. Data is the basis for the training of artificial intelligence and advanced machine learning algorithms, and data is driving the "autonomous" development of machines.

Data-driven technologies underpin the core and economic functions of modern society, and with the rollout of global 5G networks, global information networks will generate, collect, process and store massive amounts of data. Therefore, it would be wise for Canada’s Department of National Defense and the military to elevate data to a national asset, critical to both economic growth and Canada’s defence. Protecting and utilizing data means rethinking today's centralized digital infrastructure. Data security in the Internet age should be both decentralized and aggregated to avoid the risk of centralized system vulnerabilities.

4.3 Promote the development of autonomous weapons systems

The weaponization of artificial intelligence has intensified the global arms race, and it may reshape Canada’s defense strategy. Currently, there are huge advances in military system automation, equipment maintenance and surveillance, and the deployment of drones and robots due to the introduction of artificial intelligence. The United States, Russia, Israel and other countries are studying the embedding of artificial intelligence into network security and robotic systems that support combat simulation and data processing. Advanced logistics support, semi-autonomous driving, intelligent supply chain management and predictable maintenance systems represent current military applications of artificial intelligence.

Autonomous weapons do not require human participation and can carry out target identification, strike and destruction activities on land, sea, air, space and network. It is based on a combination of sensor systems that monitor the surrounding environment, artificial intelligence systems that identify potential targets and decide whether to launch an attack, and weapons capable of destroying the target. In the conflicts between Armenia and Azerbaijan, autonomous and semi-autonomous drones were used to disrupt conventional military systems, resulting in the direct failure of a series of military platforms. Recent attacks on Saudi Arabia's national oil processing facilities also confirm the increasing use of military drones in a variety of battlefield environments.

As autonomous weapons systems and data-driven technologies mature and become more widespread, they may provide state and non-state actors with the platform and tools to apply artificial intelligence and machine learning in new and disruptive ways.

4.4 Build an advanced network platform

For many NATO countries, network platforms are crucial to multi-territory operations. Web platforms make it possible to visualize and coordinate resources in complex environments. With the support of 5G and cloud computing, information systems can effectively collect, transmit and process large amounts of battlefield data and provide real-time data analysis.

Device interconnection is critical for coordinating air strikes, piloting drones, real-time battlefield space analysis, and managing highly complex supply chains. From strategy and communications to logistics and intelligence, digital platforms have become the basis for commanding complex military operations, and their data are the lifeblood of all combat fields.

In the digital battlefield space, every officer, soldier, platform and resource is a node in a complex military network. Starting with the network-centric U.S. military operations of the 1990s, digital technologies have been the foundation of advanced weapons, tactics, and strategies. From battlefield situational awareness and autonomous drones to precision-guided munitions and machine-driven psychological warfare, networks are bringing warfare into the cyber age.

Artificial intelligence is essentially a "bottom-up" technology that relies on the continuous "input" of large amounts of data to support machine learning as a "learning engine". As digital ecosystems proliferate, the network platforms and data management systems they rely on become critical to managing ever-expanding resources and people.

The highly decentralized verification system provided by DLT can eliminate possible failures of centralized nodes while ensuring that all communications and data transmissions are not attacked by adversaries. Therefore, the Canadian Department of Defense should rely on DLT, such as blockchain, to accelerate the digital transformation of the Canadian military. Overcome the inherent limitations and fragility of the original system by distributing data horizontally in a decentralized network.

4.5 Promote the rapid development of drone swarms and robots

With the rapid advancement of artificial intelligence in military applications, many countries have made great progress in the deployment of drones and robots. Among them, the development of military drones in the United States and Israel is represented. The U.S. military has a complete range of military drones with advanced technology and a wide range of uses. They mainly include unmanned reconnaissance drones, integrated reconnaissance and attack drones, decoy drones and cargo drones, which are used for battlefield surveillance, communication interruption, and military strikes. activities, etc., and its global market share occupies a leading position.

Drone swarm technology can be used for micro, small drones and unmanned aerial vehicles (UAV) to make autonomous decisions based on shared information. Modern military drones are already capable of locating, identifying and attacking targets without humans. Drone "swarm technology" can enable hundreds of drones to collect information from the battlefield to provide support for various weapon systems. Facial recognition and decision-making algorithms enable both state and non-state actors to use lethal autonomous weapons systems to carry out targeted killing missions. Equipping thousands of drones with explosive warheads can defeat air defenses and attack infrastructure, cities, military Base etc.

4.6 The emergence of “mosaic warfare”

The military threat of drones is unstoppable, and cyberattacks on critical infrastructure occur from time to time. In order to cope with the changing environment, DARPA proposed “mosaic warfare” war" concept.

The core idea of ​​"mosaic warfare" is to use cheap and flexible modular systems to deal with highly complex networked environments, in which personal combat platforms can be designed to be configurable and use digital networks to speed up dynamic response. In the "AlphaDogfight" challenge held by DARPA (2019-2020), an advanced F-16 flight simulator was used to pit computers against experienced pilots. As a result, the pilots could not match the autonomy of artificial intelligence. Attack and accuracy.

In "mosaic warfare," artificial intelligence, drones, sensors, data, and people combine to provide combat commanders with intelligence, resources, and logistical support. Modular systems demonstrate that future warfare will increasingly make use of computing, data analysis and algorithms. Artificial intelligence systems will drive highly changing and unpredictable battlefield environments and accelerate the war process.

4.7 Adversarial attacks are inevitable

The weaponization of artificial intelligence has also triggered new strategies and new methods for artificial intelligence systems. Just as cyber operations can make a computer network or machine behave in a certain way, adversaries can use the same tactics against artificial intelligence systems. This process is called adversarial machine learning, which attempts to identify weaknesses in machine learning models and correct them. use. Attacks can occur during the development or deployment phase and include misleading the model by providing deceptive "inputs" or targeting the model itself. As artificial intelligence systems become more prevalent, adversarial attacks will become increasingly attractive. Additionally, attackers often modify training or test data by creating adversarial examples that are deliberately “scrambled” or modified to cause data errors. In terms of national security, adversaries may attempt to use the same technology to influence weapons systems. If it were an isolated incident, it would likely be resolved quickly. If it occurs frequently over a period of time, it may pose a huge challenge to the intelligence collection system and affect its trust.

High-value AI systems are not the only targets of adversaries, which include biometrics and fake biometrics being exploited to impersonate legitimate users. In speech recognition, attackers compromise system and computer security by adding low-level noise. Currently, the Canadian Department of Defense has deployed "voice assistants" on its warships, hoping to improve combat efficiency by deploying artificial intelligence systems.

5. Strengthen the global governance of artificial intelligence

From drones, human-machine dialogue to military decision-making, artificial intelligence technology can double combat effectiveness. The speed and scope of data-driven warfare indicate that we are entering a new era in which lethal autonomous weapons systems will dramatically alter the global balance of power. As low-Earth orbit increasingly becomes a combat environment for military surveillance, remote sensing, communications, data processing, and ballistic missiles, artificial intelligence weaponization and space weaponization are also intertwined. Artificial intelligence, low-Earth orbit, and autonomous weapons systems represent an important turning point for global security. Researchers around the world have expressed concern about the threats they pose. They believe that if there is a lack of international consensus on the regulations for the application and development of artificial intelligence, it may lead to the emergence of crisis.

As a result, legal treaties on artificial intelligence and other digital technologies will shape the contours of war and conflict for decades to come. As the militarization of artificial intelligence develops, structuring legal treaties will be critical to reducing future conflicts. Currently, European countries are calling on EU member states to develop strategies for the use of new artificial intelligence technologies, and the United States invites allies to discuss legal issues regarding the use of artificial intelligence. NATO is launching a process to encourage member states to reach an agreement, while recognizing the profound impact of emerging technologies such as artificial intelligence on global security, and launched the Emerging and Disruptive Technologies (EDT) Roadmap in December 2019 . Canada and its allies are seeking opportunities to promote, engage and build collaboration to develop a basic framework to support artificial intelligence and other emerging technologies. United Nations Secretary-General António Guterres has also highlighted the risks and opportunities of artificial intelligence and other digital technologies and called for protective laws.

Artificial intelligence is a field of technology that impacts commercial and military applications. Given the conceptual ambiguity and political obstacles to comprehensive regulation of AI, governance issues will remain a difficult challenge for a long time to come.

6. Conclusion

Artificial intelligence has developed from a mysterious academic field to a powerful driving force for social and economic transformation. Artificial intelligence brings together advanced data, algorithms and computing power to provide the military with secure, timely and accurate information.

If Canada wants to build a military suitable for the digital age, the government, industry and academia must cooperate in an integrated manner to establish a sound innovative ecosystem. In addition to vigorously developing emerging technologies, the Canadian government and military also need to balance the changing geopolitical landscape and use information sharing, expert meetings and multilateral dialogue to strengthen international cooperation.

The above is the detailed content of The impact of artificial intelligence on military defense and security. For more information, please follow other related articles on the PHP Chinese website!