AI technology is a worldwide phenomenon, and every nation aims of being at the forefront of creating and using this technology, whereas in reality few nations truly are. Reports of alleged use of Artificial Intelligence (A.I.) in United States’ operation in Iran causing in latter’s leadership decapitation, mark a point of fundamental redefinition of the character of modern warfare. Recent news articles have reported the use of AI to process and analyse vast amounts of data, including intercepts, satellite imagery and signals intelligence, to generate summaries, threat evaluation and situational insights. Along with these, AI is also being used in target identification; it helps to locate, prioritise, cross-reference and confirm high-value targets. As it is recently used to identify Iranian leadership positions, military assets and strategic sites. This recent usage rings alarm bells as conventional strategic doctrine has integrated the use of AI. Revolution in Military Affairs (R.M.A.) has therefore received an upgrade. Cyberwar has been a looming threat, but along with it, AI now seems to have crept into traditional warfare doctrines. For example, if we study the well-known OODA loop in terms of contemporary practices and tactics, then how shall it fare if AI is integrated into the model of Observe-Orient-Decide-Act developed by John Boyd?
The OODA loop serves as the cognitive architecture for engagement. Victory is secured by the actor capable of observing with clarity, orienting with precision, deciding with coherence, and acting before an adversary achieves stability. AI functions primarily as an accelerator within the Observe and Orient phases. It has to be pointed out that, Intelligence, Surveillance, and Reconnaissance (ISR) in the military refers to the synchronized collection, processing, and distribution of data from sensors, satellites, and human intelligence to provide real-time battlefield awareness and evidently AI has been integrated by the Israeli army in the ISR chain.1
Current research on intelligent battlefield awareness indicates that machine learning models effectively fuse and analyse multi-source data while providing predictive modelling of adversary movements.2 This evolution shifts warfare from episodic awareness to a state of persistent battlespace cognition (i.e. continuous, uninterrupted, all-weather knowledge).
Across major powers, AI is being embedded in:
- ISR automation
- command-and-control
- drone swarms
- predictive logistics
- cyber operations
A 2025 defence networking survey finds AI is reshaping tactical communications by enabling real-time situational awareness, adaptive signal processing, and multi-agent coordination across domains.2
Now, if we follow the traditional OODA loop, then in the Observe phase traditional bottlenecks were, ISR overload, slow imagery analysis, and human cognitive limits. Programs like Project Maven3 demonstrated that integration is not only possible in automated video analysis, object detection in drone feeds, and rapid anomaly detection, but also that it reduces image analysis time by 70–90%, which increases detection rates in ISR streams, enabling persistent surveillance at scale and effectively expanding battlefield awareness. AI can process surveillance far faster than human analysts, enabling persistent monitoring. Therefore, the observe phase transforms from an episodic sensing into a constant all-encompassing knowledge of the theatre. Approximately 2,500 years ago, Sun Tzu valued a “perfect knowledge” of the enemy. Today with such continuous synthesis and analysis of data, perfect knowledge is the ability to realise reality in real time and predict future actions with the help of technology.
In the Orient phase, John Boyd mentions “schwerpunkt”, where orientation is not just focus or simple analysis. It includes mental models, threat perception, pattern synthesis, and predictive understanding. In this phase, modern military AI can fuse multi-domain data, detect weak signals, run predictive battlefield models, and identify hidden correlations. AI-enabled sensor fusion dramatically improves common operational picture (COP) accuracy; machine learning improve target classification confidence and predictive analytics improves maintenance and logistics forecasting. AI is performing what Boyd mentioned as “implicit guidance and control”, and under human control it could be destabilising if it is mismanaged. AI diminishes the possibility of mismanagement during this phase and helps schematize several possible courses of action.
In the decide phase, militaries still use human cognition and decision making, and credibility will degrade if a “kill switch” is made accessible to an AI. Even though several systems are automated for neutralisation of rogue drones, for example India’s ‘Indrajaal ranger’5, but mass killing is not yet under the decision of any AI as popular media report is yet
to indicate any such incident. It only helps in simulating and providing ‘courses of action’ suggestions to armed forces. There also exists a risk of automation bias6 , i.e. overuse of AI transforms into over trust on machine decisions, which shall possibly turn a battlefield scenario into a pantomime of non-human decisions. Therefore, risk modelling simulations are presently a safe course of action, with a human in the decision-making loop, for forces around the world to remain credible.
For the act phase, AI is increasingly embedded in drone swarms (for example, the Seshnaag-150 is being manufactured by an Indian startup similar to the US’ LUCAS or Iran’s Shahed-136/ Geran-2), autonomous navigation, electronic warfare response, and missile guidance optimization. Also, India has already fielded operational AI-enabled systems such as the Akashteer air defence system and the Saksham anti-drone grid.
It is not merely an acceleration of the pre-existing strategic frameworks, but rather a transformative strategic doctrine in practice, as seen in the case of the Joint-Domain Command and Control (JADC2) of the USA. The basic purpose is to connect sensors with shooters across domains and create a unified battlespace network. The command and control receive assistance in data fusion, targeting recommendations and cross-domain coordination.
Another example would be the newly developed idea of Mosaic warfare by Defense Advanced Research Projects Agency (DARPA), USA, where an asymmetrically large volume and a variety of weaponry and platforms are used in a network during an attack. In such a network, the use of AI is integral because the communication among each unit of the force structure is crucial. Also, the AI helps keep the mission adaptive due to its capability of real-time sensing and data processing. Therefore, this shows that traditional strategic doctrines such as the OODA loop can integrate AI with ease and be subject to modernisation. This widens the scope of conventional strategic tactics and ideas and opens up the possibility of their continued usage in strategic thinking and warfare worldwide.
We can with ease say, that a transformation is taking place From human paced warfare, to a new form of machine paced competition ,where the speed of their systems can dictate the resultant victory or loss. Therefore, modern war is becoming a competition of feedback loops and the processing speed of computers. In a new multipolar world, powers of all sizes desire to harness the uses of AI to grow in aspects of security that are still relatively new to the world. The future is also uncertain as the existing conventions on cybercrime (e.g. Budapest 2001) struggles to keep the cyberspace in check, and yet new frameworks such as the EU Artificial intelligence Act try to build a new legal framework for AI in international law. But using it in war, even though not directly to fight it, but in assessment, synthesis of data, simulation and making predictions, i.e. using AI as a helper has pushed the intelligence, surveillance and reconnaissance (ISR) at the edge of perfection.
Bibliography/ References:
- Kurek, Julius, and Björn Laurin Kühn. “Habsora and Lavender Artificial Intelligence Systems – The Missing Piece Towards a Fully Algorithmically Automated F2T2EA Kill Chain?” EPIS Think Tank, August 19, 2024.
- ArXiv Open Research. “Intelligent Battlefield Awareness and Predictive
Modelling in Multi-Domain Operations.” Journal of Defense Tech Analysis, 2025.
- Pellerin, Cheryl. “Project Maven to Deploy Computer Algorithms to War Zone by Year’s End.” U.S. Department of Defense News, July 21, 2017.
- Pfaff, C. Anthony, and Christopher John Hickey. Integrating Artificial Intelligence and Machine Learning Technologies into Common Operating Picture and Course of Action Development. Carlisle Barracks, PA: US Army War College Press, 2025. https://press.armywarcollege.edu/monographs/980.
- WION, “India’s First Anti-Drone Patrol Vehicle: What Is Indrajaal Ranger and How Does It Stop Drones?,” November 26, 2025, .
- Lim, Nelson. “Your New Teammate Is a Machine. Are You Ready?” RAND Corporation. October 14, 2025.
- Press Information Bureau (PIB), “Akashteer: The Unseen Force Behind India’s New War Capability,” May 16, 2025, .
- Kaul, Aditya Raj. “Army Launches Indigenous ‘SAKSHAM’ Anti-Drone Grid To Boost Air Defence.” NDTV. October 9, 2025.
- Stew Magnuson, “DARPA Tiles Together a Vision of Mosaic Warfare,” DARPA: 60 Years of Innovation, 2018, 22,
- Reuters. “US Used B-2 Bombers, Suicide Drones, Anthropic AI in Strikes Against Iran: Report.” March 1, 2026. https://www.reuters.com/world/middle-east/us-strikes-iran-b-2-bombers-ai-drones-2026-03-01/.
- Dwoskin, Elizabeth. “Anthropic’s AI Tool Claude Central to U.S. Campaign in Iran, Amid a Bitter Feud.” Washington Post. March 4, 2026. https://www.washingtonpost.com/technology/2026/03/04/anthropic-ai-iran-campaign/.
- Hoffman, Frank. “Conflict in the 21st century: the rise of hybrid Wars.” Potomac Institute for Policy Studies, Dec. 2007, p. 58, https://www.potomacinstitute.org/images/stories/publications/potomac_hybridwar_0108.pdf.
- The decision lab, “ The OODA loop”, https://thedecisionlab.com/reference-guide/computer-science/the-ooda-loop.
The views and opinions expressed in this post are solely those of the author and do not necessarily reflect the views, positions, or policies of the Jadavpur Association of International Relations (JAIR) or any of their affiliates. JAIR Learning Commons serves as a platform for academic learning and student expression and encourages diverse perspectives and critical engagement.
About The Author
