On May 10, 2025, Pakistan and India announced a ceasefire bringing to an end a confrontation which militarily did not last for more than four days and nights. Such conflicts are categorised as limited wars (as opposed to a general war), in which the aim, scope, rules of engagement, intensity, duration, geographical span, etc. are scaled down. Neither side seeks decisive results and both aim to end the engagement in a position of military strength that yields a political advantage during negotiations. Therefore, a limited war is controlled, calculated and often influenced by diplomatic concerns.

Within this very broad definition of a limited war, we will look in a little more detail at the ‘May War’ of 2025. The conflict was ignited by an attack on April 22 by Kashmiri militants in Pahalgam resulting in the deaths of 26 civilians tourists. Within hours India attributed the attack to a little known group it alleged was affiliated with Lashkar-e-Taiba, leading to heightened tensions.

This may or may not have been a false flag operation by the Research and Analysis Wing (RAW), the Indian equivalent of United States’ Central Intelligence Agency (CIA) or Israel’s Mossad, etc. Was it a coincidence that it occurred during a visit by the American vice president just as a terrorist attack on Sikhs, allegedly carried out by Kashmiri militants, occurred during former US president Bill Clinton's visit in 2000? It reminds me of a quote attributed to Ian Fleming's James Bond novel, Goldfinger: "Once is happenstance, twice is coincidence, three times is enemy action". In her autobiography, Clinton’s foreign secretary Madeline Albright blamed the killing on Hindu militants (for Hindu Militants read the Rashtriya Swayamsevak Sangh, the militant wing of the ruling Bharatiya Janata Party).

Attacks by militants in Indian Held Kashmir (IHK) have been occurring periodically and even if Pahalgam was not a false flag operation, in such an eventuality, India was prepared to execute a contingency that would gave it political and military gains. The instant gain was finding it an excuse to suspend the 64-year-old Indus Basin Water Treaty that had survived two wars and withstood many twists and turns. Having what they considered a political victory, the Indians now wanted to militarily ‘punish’ Pakistan for supporting terrorism in IHK.

I have been on the Directing Staff at the National Defence University and have taught all manners of conflict strategies relevant to the scenarios in the Subcontinent, including limited wars. However, it was during the Kargil conflict that I realised that in many ways a limited war is more difficult to manage. In a general war troops are mobilised, formations and units deployed in the battle zone according to a predetermined plan, air bases placed on war footing, satellite bases activated, ships and submarines move out to sea, etc. for a no holds barred conflict.

On the other hand, in a limited war mobilisation is very selective and every deployment and action has to be gauged against counteractions by the enemy. A limited war only remains limited if both sides choose to do so and if one side is reckless, the escalation ladder becomes steep.

After the Pahalgam incident on April 22, Indian security forces spent two days conducting joint operations in Jammu and Kashmir, which they allege led to the death of a senior Lashkar-e-Taiba commander. They also demolished residences they claimed belonged to suspected militants. Concurrently they activated the Line of Control (LoC) with both sides exchanging artillery and small arms fire across multiple sectors. By April 28, diplomacy gained traction to limited avail with the US seemingly sitting on the fence. Finally on the night between May 6 and 7, India launched "Operation Sindoor," with missile strikes on what they claimed was terrorist infrastructure in Pakistan and Azad Jammu and Kashmir.

This is when the conflict got interesting, but to appreciate the scale and nature of the PAF response we have to go back to a similar incident in 2019 when the IAF struck what it alleged was a terrorist camp near Balakot. Operation Swift Retort, the PAF response to India's strikes showed glimpses of its prowess in the electromagnetic sphere. The counterstrike package of the PAF had a very strong component of electronic warfare that enabled their aircrafts to engage the enemy without activating their own radars thus concealing their locations.

But back then, the PAF was still far from possessing the technology need to excel in multi-domain operations that encompasses real-time imagery, cyber units disabling enemy radar, drones providing surveillance and fighter jets launching precision strikes – all while being coordinated through secure, networked communications. The PAF had acquired radars from different countries and integration of those technologies of different origins was achieved by an indigenous Data Link that integrated all these systems. According to military analysts, the PAF decided to build on this capability and under a modernisation plan, the PAF established the PAF Cyber Command (PAFCC) and the PAF Space Command (PAFSC) in Islamabad. It also created the NASTP (National Aerospace Science & Technology Park) which has become a hub for various indigenous programs.

One of the biggest advantages that the armed forces of Pakistan have over the Indians is that development, testing and manufacture of weapons and military systems is guided by the services themselves. Within a few years, and of course with much assistance from the Chinese, the PAF developed a capability which took aerial combat to a completely new level. EW was no longer just a force multiplier and coupled with AI, it became central to a new art of warfare that I call AI-Centric Cognitive Warfare where ‘AI-Centric’ highlights the primacy of machine intelligence and ‘Cognitive Warfare’ indicates the strategic aim of winning the battle. In other words, war by disruption, not just destruction

Parallels can be drawn from history where dramatic and unexpected innovations changed the method of warfare. The feudal knights in armour (the heavy cavalry) who had dominated the battlefields of medieval Europe were defeated by the arrows of the English longbows which penetrated their plate armour. The advent of large cannon in late medieval period was a watershed moment that sounded the death knell of siege warfare by rendering traditional castles and city walls useless. The German Blitzkrieg in the opening stages of Second World War marked a dramatic shift from the static trench warfare of the previous World War and its core principles continue to influence military doctrine till now. Each of these breakthroughs was more than a tactical improvement — they represented a fundamental change in how wars were fought. AI-Centric Cognitive Warfare may prove to be the next such transformation, reshaping battlefields with speed, precision, and control over information itself.

In future limited wars (as well as operations on a larger scale), AICCW will shift the battlefield focus from brute force to brainpower. The aim will be to deliver maximum damage quickly — not just by destroying enemy assets, but by breaking their ability to think, respond, or coordinate effectively. AI will take the lead in planning and executing operations with speed and precision. It identifies weak spots in the enemy’s defences, disrupts their communication, and creates confusion in command structures — often before the first missile is even launched. Weapons of destruction like tanks, fighter aircrafts, missiles, artillery, drones, etc would still used, but their employment will be guided by AI systems that target not just hardware, but the minds and systems behind them.

A limited war fought within the concept of AICCW would open with launching cyber and electronic attacks to jam radars, blind sensors, and distort battlefield awareness. This would be in tandem with strikes to overwhelm the enemy’s decision-making and disrupt leadership and command structure by neutralising communication hubs, and information nodes (e.g., radio trucks, SIGINT stations), making it hard for the enemy to fight back. Having established a secure battlespace eliminate or cripple the enemies’ assets identified through AI analysing enemy patterns, communication flows, and sensor emissions to target command elements. Rather than “kill everything,” own forces will be able to focus on the neutralisation/destruction of the most critical assets. As the battle unfolds, AI will adjust the operation in real time — redirecting drones, retargeting missiles, and shifting tactics. There are indications that the Pakistan Army is also developing an AICCW capability. It has officially announced that the Corps of Signals has been elevated from a supporting arm to a combat arm. This signifies a shift in the role and responsibilities of the Corps, placing them at the forefront of military operations rather than simply providing support and is a noteworthy development in the Pakistan Army's structure and operational capabilities. I would not be surprised if our ‘Silent Service’, the Pakistan Navy is already well ahead in developing an AICCW capability.

In a region like South Asia where tensions can spike quickly and escalation must be avoided, AICCW offers a way to achieve decisive results without a full-scale war. These AI-led limited wars could be over in days, with one side effectively paralysed before it can respond. The key is to win the cognitive battle — to confuse, deceive, and disable the enemy mentally and operationally before they can regain control. However, such fast, AI-driven warfare comes with dangers. Misjudging the enemy’s red lines or failing to control escalation could turn a limited war into a larger conflict. And as machines take over more decision-making, the role of human judgment and restraint becomes more critical than ever.

Author’s Note: all material for this article has been gathered from open sources.

Syed Ali Hamid is a retired Pakistan Army major general and a military historian. He can be contacted at syedali4955@gmail.com

All facts and information are the responsibility of the writer