Today, AI is playing a momentous role in defense and is being used in a gamut of applications across different domains. From enhancing situational awareness by processing vast amounts of data from sensors, now defense systems provide real-time intelligence, and can detect threats with increased accuracy. Algorithms power autonomous systems, like drones and unmanned tanks & other vehicles, enabling them to perform reconnaissance, surveillance, and combat operations with reduced human intervention. These algorithms are also used for threat detection, quick decision-making and operational efficiency – which has altered the landscape of the new battlefield. Autonomous defense systems are increasingly being used in air, ground and water combat. UAV’s rely on machine learning algorithms; that are learning algorithms letting systems learn on their own from data and make predictions, without any special programming.
ML algorithms are broadly classified into supervised, unsupervised, and reinforcement learning approaches. Linear regression in supervised learning helps fast acquire the relation between dependent and independent variables. AI and ML algorithms together provide enhanced cyber security in defense applications to analyze data traffic, detect and swiftly mitigate cyber threats, spot vulnerabilities and save against cyber-attacks. AI-powered systems also aid in quick response and help form a sheath of defense against emerging threats.
In recent years, neural networks and deep learning algorithms have turned out as a benchmark of modern artificial intelligence, letting machines to recognize patterns, take decisions, and solve complex problems with human-like performance. Inspired by structure and function of man’s brain, neural networks are computational models have evolved dramatically over the past decades, driven by advances in data availability, computing power, and algorithmic innovation.
Networks containing multiple hidden layers are called deep neural networks which come in types. The output layer produces the final prediction, which might be a classification or a constant value. Common network architectures include FNN’s – Feedforward Neural Networks where information flows in one direction from input to output without cycles and CNN’s – Convolutional Neural Networks, that are designed for processing grid-like data such as images. CNN uses convolutional layers to automatically detect spatial hierarchies of features, like edges, textures and objects.
While RNN’s are more suited for sequential data like time series or natural language. They maintain a hidden state that captures information about previous inputs, enabling memory of past context. – Transformers: a more recent architecture that relies on self attention mechanisms to process sequences in parallel, overcoming limitations of RNNs in long range dependency modeling. Transformers power many state of the art language models. These are just a few of neural networks.
Experiments in AI had been ongoing since many decades. According to a report published by Published by Nonviolence International Southeast Asia, a Philippines based think-tank has observed that, since the Cold War, governments have been experimenting with weapons systems that have been programmed with increasingly sophisticated AI functions. LAWS, i.e Lethal Autonomous Weapons Systems (LAWS), as they are called gained specific impetus owing to prominent and aggressive developments in advanced weaponry. “Speed and efficiency are commonly identified as the primary benefits of equipping weapons systems with AI technology,” cites the report.
The US- Iran conflict has witnessed high usage of AI, systems integrated with cognitive and cyber technologies. There are an array of predictive analyses tools which can effectively track and trace enemy activity, their hideouts and possibilities of what the next move is all about! The modern battlefield is all about tasks which target information, and figuring out what the human mind will likely pursue. Military action follows thereafter, which is believed to be more precise & accurate. In this fight of cognition, sets of data are collated and transformed into what could be the adversary’s next move. Many AI generated images are close to real, and leave human minds speculating. Social media is afloat with Israel’s Prime Minister’s conspicuity.
Together, it can be deduced that the increasing use of AI-enabled technologies, coupled with social media are creating warfronts. Two countries use two opposing factions use artificial intelligence to create content – deemed as deepfake. This content then spreads across social media platforms to celebrate military successes and intensify the internal conflict. Lost for right information, civilians remain clueless for long, making it difficult for civilians to interpret ground reality. The effects of cognitive warfare isn’t restricted to the war field, its ripples cut across political and social organizations, who if don’t act responsibly can extend wars that are based on assumptions that are not human. AI platforms are used in combination of different fundamentals of cognitive warfare, both for self defense and attacks.
While both neuroscience and neuro-technologies have evolved as tools for influencing the mind by affecting the brain. Today, AI platforms have turned into the principal means of influencing the mind by stage-managing the information ecosystem. It has a rate which goes beyond human capabilities and certainly isn’t immune to human error. AI has already been used in weapons systems such as active protection systems, drones and systems that prevent missiles and projectiles from destroying a target be they tanks or sentinel robots. While governments claim that most of these weapons systems are used to intercept and eliminate incoming projectiles, the global concern over the heavy usage of artificial intelligence is growing with every confrontation. Researchers and military contractors collaborating over AI based weaponry has got civil society rightly worried.
Though AI has advantages like use in training and simulation to create realistic virtual environments and intelligent opponents, revolutionizing the way military is prepared for real world military operations and simulating real-world scenarios allowing defense personnel to train, plan, and evaluate strategies in a safe and cost-effective manner, it’s disadvantages cannot be ignored that are lack of human judgment and intuition, vulnerability to cyber attacks and excessive dependence on technology and infrastructure. It is important that such technologies don’t get to the hands of the wrong actors. Prudence and vigilance are not mutually exclusive to one another.
