The application of Artificial Intelligence (AI) and Machine Learning (ML) is increasing by the day in the armed forces. In fact, the armed forces are gradually becoming the greater proponents of ML & AI. However, a host of grey areas remain that put the focus on what are the ethics in these fields to ensure we are not face-to-face with a coup by our intelligent machines. The author provides an overview of the wider dimensions of use of ML & AI in both operations and logistics. He also cautions about possible unbridled enhancements and progression in both the areas.
Artificial Intelligence, Revolutionising Warfare
“The bravest are surely those who have the clearest vision of what is before them, glory and danger alike, and yet notwithstanding, go out to meet it” – Thucydides
Whilst, the proponents of the expression about the brave, explicitly mention of a clear vision of what is before them, today the definition of braves is undergoing a metamorphosis. There is nothing in front; there is nothing to the sides. The battle space is amorphous and the warrior an armchair technologist with algorithms on his palm. Both glory and danger operate within the six-inch precincts of his temples! Welcome to the fourth Industrial Revolution, that of ‘Artificial Intelligence’ (AI).
Demystifying the Clutter
Google and Amazon, use AI and Machine Learning (ML) to anticipate customer behaviour using a blend of both along an ‘AI powered forecasting analytics’. It is important to distinguish between the two. As per Tom M Mitchell, Professor and former chair of ML Department at the Carnegie Mellon University, ML a branch of AI, is the study of computer algorithms that improve automatically through experience. Whilst, it is the self-propagating ease with which business is articulated today due to ML, there is also the inherent danger of handing over our lives to the alter of these machines, as the Internet of Things (IoT) is beginning to exhibit.
According to Techopedia, a web-based portal, “AI is a branch of computer science that aims to create intelligent machines”. Therein also lies the challenge, since the machine would have to be programmed for characteristics, hither-to-fore associated with only human intelligence.
Since ML is associated with examining large data sets and comparing its behaviour, there are subsets to ML that get into the realm of ‘deep-learning’ and ‘neural networks’. Cognitive computing, Predictive application Programming Interfaces (APIs), Natural Language Processing (NLP), Image Recognition and Speech Recognition to data sets that can be processed by NLP are all in the realm of the present day reality. It is from these commercial applications that the defence industry has picked up the threads, using its guile and imaginative application models to sketch a future on the throes of revolutionizing warfare!
Recall how in 1997, Garry Kasparov, then the world chess champion, had to face defeat at the alter of ‘Deep Blue’, a super computer from IBM using ‘Tree-search’ algorithms from AI. These terminologies are a science onto themselves and out of the preview of this paper.
The intelligent assistants, Google home, Alexa and Siri and ML powered Netflix and Amazon Prime have become facilitators in our daily lives. Obviously, with such varied application, the utility of AI transcends diverse fields like education, manufacture, financial risk, health care and defence, amongst others.
The Defence Muse
Prime Minister Narendra Modi at the Def Expo 2018 mentioned the use of AI and Robotics in Defence. He further went on to state, “New and emerging technologies like AI and Robotics will perhaps be the most important determinants of defensive and offensive capabilities for any defence force in the future”. In 2018, the government published two AI roadmaps – the Report of Task Force on Artificial Intelligence by the AI Task Force constituted by the Ministry of Commerce and Industryand the National Strategy for Artificial Intelligence by Niti Aayog.The Task Force constituted in February 2018, with a multi-stakeholder composition under the Chairmanship of Natarajan Chandrasekaran, Chairman, TATA Sons, hasmembers from the Industry, Professionals, Academia, Services and the Government.
The Task Force handed over its report to the Defence Minister Nirmala Sitharaman on 30 June 2018. Some of the recommendations include keeping defence needs central, making India a significant power for AI in Defence including chemical, biological, radiological and nuclear warfare, working with start-ups in the commercial sector on a ‘partnership model’ with its funding and resource base availability. The same model is to be adopted by the military subsequently, thus benefiting the Defence industrial base. The Government has also initiated the establishment of the National AI Mission (N-AIM) with a fund allocation of Rs 1200 crore over five years.
Militarily, AI and ML have manifested in the development and use of Drones, Unmanned Aerial Vehicles (UAV), Unmanned Ground Vehicles (UGV), semi-autonomous and autonomous vehicles, unattended ground and aerial sensors. Lately, Swarm technology, Predictive Analytics has stepped up the hyperbole of utility and suddenly, the options seem endless. The military decision-making process has been enhanced and acquisition of actionable intelligence accelerated through speed and quality of Intelligence, Surveillance and Reconnaissance (ISR), realising real-time results.
AI is fast transcending cyberspace, geo-space and space domains. The exponential possibilities that started with commercialization are shifting to AI chips manifesting in automated weapon systems, autonomous vehicles to include tanks, warships, antisubmarine weaponry, drones, robots, third generation plus missiles and as per the RAND Corporation report- transcending even the nuclear domain. A report leaked in 2015, mentioned of Russia having developed a “nuclear-armed undersea autonomous torpedo named Status-6 Oceanic Multipurpose System with an enormous thermo-nuclear warhead.”
India’s manifestation of AI in defence has taken the following form:-
- Robotics: India has balanced the operational and logistics requirements. From small robots that diffuse mines and IED to large and heavy autonomous vehicles grouped as a vanguard for the frontline manned vehicles.
- Target Recognition & Identification: AI would calculate move rate of the targets and fix gyroscope and accelerometer movement and facilitate in delivering precision fires. It would factor the error and biases due to altitude, temperature, air density, pressure, crosswind, and barrel wear etc., enhancing strike rates. It could pre-empt movement of own forces on adversarial axes, thereby, denying critical spaces in manoeuvre warfare.
- Military Platform: The entire gamut of a weapons platform would get enhanced in efficiency, reduce battle fatigue and commensurately have better life cycle management.
- Cyber Security: AI facilitates getting on top of the cybersecurity loop. Any breach can be forewarned and identified for real-time detection and response. The attackers could be evicted before causing any serious damage. It is a powerful weapon against the puissant cyber-attacks and trespasses. With the help of AI, bio-metric logins, detection of physical features like retina scans, fingerprints etc., make the security system formidable.
- Operational Logistics:AI would help in ‘Augmented real-time decision making’, considering the calculation of load tonnages per vehicle, carrier selection, time for loading, movement and delivery. ‘Predictive Analysis’, would help in vehicle inventory availability, maintenance task, preventive maintenance in real-time facilitating ‘Strategic Optimisation’, to include the availability of drivers, vehicles availability for loads, mode of transportation, transfer of bulk, is all facilitated by AI.
- Threat Analysis and Situation Awareness:Through the use of highly specialised advanced technology platforms and by use of cognitive computing and data mining, ML and natural language processes, complex situations with ambiguity and uncertainty can be analysed accurately.
- Logistics Model:Battlefield casualty evacuation process, turnaround time, mode of evacuation depending on the severity, availability of critical drugs, shelf life and storage, vehicle casualty and repair echelon statistics, spare parts availability, the lead time for availability of major assemblies etc. AI would orchestrate the rollout plan in a precise and ‘Just in Time’ manner without the necessity of reserves at all levels leading to lean logistics models.
- Disaster Management:The National Disaster Management Authority (NDMA) utilises AI, Data analytics and deep learning for disaster relief in times of natural calamities. Work is in hand to incorporate predictive analytics to anticipate with a fair amount of cushion time to marshal resources, much like the Tsunami warning in the coastal areas.
In May 2018, the Secretary, Defence Production, Ajay Kumar had stated unequivocally, “It is India’s preparation for next-generation warfare. This (AI) is where the future is going to be. We need to prepare ourselves for the next generation warfare which will be more and more technology driven, more and more automated and robotized.”
The offensive and defensive capabilities of AI are phenomenal. From infecting networks with malware in milliseconds to spoofing frequencies and jamming radars, the military has a task at hand to deal with these new shenanigans. On the defensive side, ML would facilitate warnings of an impending attack through predictive analytics, combating spam and detecting malware.
Of late, big names like Stephen Hawking, Elon Musk, Steve Wozniak and Bill Gates have expressed concern on the risks posed by AI. The Puerto Rico Conference in 2015 deliberated on aspects like ‘How to tackle criminal AI; AI in the hands of non-state actors for ransom; what if any, are the AI ethics; what are the legal frameworks for military systems & financial systems to be operating in the AI domain’.
A candid scenario algorithm did reveal the possibility of complete machine control, reflected in many Sci-fi movies. A segment is working consciously on the anti-thesis;whereina collective conscious effort is underway to keep the machines at a threshold, guarding against the possibility of a coup.
Sumathy Ramesh, in his reply to Max Tegmark, a professor from MIT on his article, ‘Benefits and Risks of AI’ in a publication of ‘Future of Life Institute,’ emphasised capping the fear of machines controlling humans. He stated, “The ray of hope I see at this stage is that Artificial Wisdom is still a few years away because human wisdom is not coded in the layer of the neutron that the technology has the capacity to map.”
Elementary, my dear Watson! As Sherlock Holmes would say in the works of Sir Arthur Conan Doyle. The Center of Gravity, from the Clausewitzs’ Trinity has shifted after all to Sun Tsu’s critical nodes and connectivity between the Army in the field and the rest of the Trinity. Has AI then destroyed the effectiveness of Operational Art?
Brig Yogesh Kapoor (Retd)
(Disclaimer: The views and opinions expressed in this article are those of the author and do not necessarily reflect the official policy or position of BharatShakti.in)
Decision making in an environment where speed of receipt information is delayed only by nano seconds from the occurrence of a security crisis and the simultaneity of other such crisis occurring, will add tremendous pressure on the Decision makers to respond swiftly.
In such a scenario, where the situation on ground is being streamed live by Media ,additional pressure will be created on the Decision Makers to Respond .This is going to increase the probability of bad decisions.
AI assisted weapons such as aircrafts , hypersonic missiles moving at five times the speed of sound,Cyber attacks at click of a mouse paralyzing major infrastructures such as Power Distribution Grids, Transportation hubs etc gives Decision Makers nano seconds to a minute at the most to counter the threat .
Incase delayed, the chaos and confusion created by the attack will breakdown social order ,with Social media adding to this chaos. This will add to the burden of the decision makers to initiate Damage Control and to restore normalcy , in addition to a speedy response.
Not only this speed compresses the time for decision making but it also disrupts traditional and more predictable patterns of conflict escalations .
The future Political and Military Leadership will have to be trained to handle this “Speed” generated by AI.
A wonderful article.