“Artificial intelligence is the future of not only Russia, but of all mankind,” and “whoever becomes the leader in this sphere will become the ruler of the world”
– Vladimir Putin, 2017
Technology has been chiefly a causatum of war and always played an un-ambivalent role in its outcome. Artificial Intelligence (AI) is one of the very few technologies whose genesis, while lying outside the military realm, has demonstrated its ability to reshape the future of warfare. In the summer of 1956, John McCarthy organised a seminar on thinking machines and coined the word ‘Artificial Intelligence’, basing it on the conjecture that a machine can simulate every aspect of learning or any feature of intelligence. Since then, AI has come a long way, exhibiting its potential to replace human performance and making militaries worldwide stand up and take notice. Whether it was the defeat of the reigning world champion of game “Go” (much more complex than Chess), or the development of an encryption system independently or achieving 99% accuracy in the diagnosis of millions of mammograms, thirty times faster, AI has not only matched but exceeded human performance and militaries across the world are in awe.
Modern battlefields are information-intensive due to the networked battlespace environment and non-military characters with unknown allegiances and intentions. To collect, analyse, and interpret the information to make out the picture of what is happening and take the best action considering the consequences of many other probable actions, all as of yesterday, make it an impossible task for any military commander. AI-supported information agility is essential, dramatically accelerating the Observe Orient Decide Act (OODA) loop. AI is the new reality, whose militarisation is on a hypersonic track, towards the quest for “superintelligence” or “Singularity” in battle space. Thus, it is no surprise that the leading military powers – the US, Russia and China- are all investing heavily to augment their current capabilities adopting AI.
However, AI cannot be perceived as an elixir, a ‘plug and play’ technology, which one can buy like any military equipment transforming the military instantly. It requires development and a bare minimum start point to exploit its potential. Networked systems, improved reconnaissance, increased situational awareness, precision-guided ammunition and integrated action, all managed by a well-trained force, are prerequisites for effective AI implementation. Even when any military has the bare minimum, like with any new technology, AI needs to cross the barrier of – reliability, adaptation and safeguards, before its adoption.
Having recognised AI’s role for future strategic dominance, the Americans released a National Artificial Intelligence Research and Development Strategic Plan in 2016 and have included AI in their military strategy as the “Third Offset”; first is nuclear deterrence and second reconnaissance and precision strike. China, not far behind, is investing heavily to transform from its today’s “information-based” warfare to “intelligence-based” warfare. Pursuing its “leapfrog development” policies and military-civil fusion strategy, it intends to catch up and overtake the US military. The success of Chinese companies, particularly Baidu. Alibaba and Tencent and a vibrant technology start-up ecosystem, have made tremendous progress in speech recognition, computer vision, and interactive systems, leaving many similar efforts in the US far behind. Having learnt from the ongoing conflicts in Syria and Ukraine, Russia is also focussing on AI and intends to achieve information dominance on the battlefield.
Indian military efforts, as far as the incorporation of AI are concerned, have been on predictable lines. The Defence AI Council (DAIC) under Defence Minister and Defence AI product Agency (DAIPA) under Secretary Defence Production have been formed, and the Centre for Artificial Intelligence and Robotics (CAIR) under Defence Research and Development Organisation (DRDO) has been tasked as the lead agency for AI in defence. While these steps are appreciative, what needs to be done is a rational outcome-based approach which focuses on improving the current situation, making it more apt for handling and inducting, AI-associated technologies.
Structural changes need to be done to handle the fast technology development cycle of AI. The Ministry of Defence (MoD) has to transform its development and acquisition process to cater to rapid development and induction. Indian military establishment has to be cognizant that there is a paradigm shift in how technologies (especially AI) are developed, monitored, funded and inducted. Solutions which were relevant decade back will not be viable in current times where the development to induction cycle is exceptionally rapid, requiring understanding and quick decisions.
By the time any committee convenes, members are explained, proposal is examined, decision taken and tender floated, the technology would be outdated. The system also needs to change to accept risks and failures in the development, trials and even sometimes post induction. “Zero error, zero loss” policy will be the biggest impediment if continued with, while handling development and induction of AI. MoD, also needs to urgently leverage commercial efforts/innovations in this field, which can be harnessed into military use and allow relevant bodies on AI to be headed by professionals rather than by senior bureaucrats or senior military personnel. Indian private software industry is bursting with competent professionals with relevant tech, however what is required is a tapping the pipeline and bringing the same to the MoD. The initiative for the same has to come from MoD and protocol for exchange of information also has to be civilian led.
Individually, each service need to incentivise and maintain personnel with high levels of technical expertise by incorporating changes in their recruitment, training and promotion policies. Effective employment of AI-enabled automated systems require highly trained and skilful personnel, which need to be motivated, rewarded and retained. The personnel policies have to be appreciative of the fact that individuals in this field might not be seeking a long-term career in the armed forces and might like to relinquish their job after a short stint. Also, few individuals would like to contribute while remaining in the civilian sector. Additionally, while in service, these personnel cannot remain detached from the commercial world since the commercial sector is the nucleus of AI-related technologies and hub of innovation, thus a regular interaction needs to be ensured. Civil and military sectors have to work in unison, with the initiative coming from the military. The future of any technology is the duality of use, which is quickly becoming apparent and should not be missed under false ego and pretence. Indian armed forces cannot step into this new world with the tools and processes of the old bygone era. It needs to transform and adapt if it wants to retain their military edge.
The approach towards AI needs to be gradual: digitisation, networking and interoperability, followed by training and process change. Extensive data collected from varied sources is the first step towards AI. Data from various inputs (voice, image, target coordinates, platform parameters, location, weather, etc.) must be digitised before processing or sharing. This massive amount of current and legacy data has to be digitised before it is mined for insight forming the foundation of the AI ambition of the Indian military. Demonstrating a swarm of drones attacking a target, developing language translating software (Mandarin to English), making robots that can play Chess, monitoring internet traffic for the search of keywords; while being start points, cannot be taken as an indicator of success.
Platforms and systems need to be networked and integrated into the command, control, communications, computers, intelligence, surveillance and reconnaissance (C4ISR) capabilities. The fusion of systems and sensors across platforms and domains is a must for an effective AI implementation. The last of the steps is interoperability, wherein an individual platform or weapon system can be part of a more significant force and perform in unison, presenting a coherent picture to the operators. Networking does not translate automatically into interoperability, and it barely indicates that the systems can communicate with each other. All this can be a significant challenge for a ‘buyer’ country like India, where most of the weapons and equipment used by its armed forces are bought from various countries and manufacturers, with varying formats, protocols and contractual conditions. However difficult, the steps are critical and should be the primary focus of armed forces and DRDO.
AI is defined as a simulation of human intelligence by machines/computers systems and includes expert systems and machine learning such as neural networks and deep learning. It includes computer vision, image recognition, language processing, cognitive learning and human-robot interaction. AI is leading towards a profound revolution in all fields, including military and Indian armed forces need to leverage the country’s leading position in the IT industry, aiming to be the ‘leader’ and not a ‘follower’. This is one path Indian armed forces cannot take in isolation and need to involve the commercial and academic sectors. Indian military needs to adapt and make significant changes in its organisation, recruitment, retention, acquisition and operation policies to remain relevant and win wars. Human – Machine integration is the future, which has moved from keeping humans “in the loop” to “on the loop”. AI is removing the “fog of war”, and the Indian military cannot afford to remain in the fog.
– The writer is voluntarily retired from active service after more than 21 years of commissioned service. He is an alumnus of Naval Academy (first course 10+2 X) and Defense Services Staff College and a specialist in Anti-Submarine Warfare. Presently, he is pursuing PhD in Defense Industrialization and Exports (India) from Indian Institute of Foreign Trade (IIFT). New Delhi. The views expressed are personal and do not necessarily reflect the views of Raksha Anirveda.
