Sense, Sense-making and the Changing Character of Warfare

War has always been a contest of wills, a clash where human leaders try to outthink and outmaneuver their enemies. For centuries, the fog of war has been a constant companion, a mix of confusion, fear, and missing information that makes every decision difficult. Soldiers and commanders have relied on their own eyes, ears, and judgment to cut through that fog, deciding what is real and what is a trick. But in the twenty-first century, a quiet revolution is changing how armies see the battlefield. Instead of relying solely on human senses, modern militaries are handing the job of seeing and understanding over to machines. Dense networks of sensors, satellites, and artificial intelligence now gather data at speeds no human could match, sorting through mountains of information to tell commanders what to shoot at and when. This shift is not just about making decisions faster; it is about changing who or what gets to decide what counts as truth in the middle of a fight.

A new paper by David Galbreath, a researcher at the University of Bath, explores what happens when we let machines do the heavy lifting of interpretation. The study asks a profound question: if a computer decides what a soldier sees, and that computer makes a mistake, who is responsible? The paper argues that these digital systems are not just tools, like a better rifle or a faster truck. Instead, they act like the invisible rules of a game, shaping what is possible, what is visible, and who holds the power. Galbreath suggests that when these systems fail, they reveal the hidden political structures of modern war, showing us that the machines themselves are now active participants in the conflict, not just passive helpers.

To understand this, one must first look at how the idea of "sense-making" has changed. In the past, a radar operator would look at a screen, see a blip, and use their training to decide if it was a friendly plane or an enemy missile. Today, artificial intelligence systems do this sorting automatically. They scan vast amounts of data, identify patterns, and flag targets for human review. This delegation of mental work means that machines are now defining the reality that human commanders experience. If the machine thinks a civilian car is a military truck, the human commander might see it as a truck. The paper points out that this creates a new kind of danger. The machines are not infallible; they can be tricked by electronic interference, or they can make mistakes because they were trained on bad data. When they fail, the results are not just technical glitches; they are crises of authority.

Galbreath draws on a theory called "mechanical constitutionalism" to explain this shift. This idea suggests that machines are not neutral objects but are deeply woven into the legal and political order of society. Just as a constitution sets the rules for how a government works, digital networks set the rules for how war is fought. They decide what information gets through, how fast it moves, and who gets to see it. When these networks work smoothly, they are invisible. But when they break—when a signal is jammed, a computer is hacked, or an algorithm misclassifies a target—the hidden rules are exposed. A breakdown in the system reveals that the power to make decisions has been transferred from human hands to code. This is a critical moment because it shows that the state's ability to wage war depends entirely on the health of these digital infrastructures. If the network collapses, the state's ability to perceive and act collapses with it.

The paper examines real-world examples to show how this plays out. In recent conflicts, such as those in Ukraine and Gaza, both sides have used artificial intelligence to locate targets and guide drones. These systems can process images and data far faster than any human team could. However, this speed comes with a cost. The paper notes that when these systems are jammed by electronic warfare or tricked by false signals, the entire chain of command can be thrown into confusion. A radar might be flooded with fake targets, making it impossible for operators to know what is real. An algorithm might mistake a harmless object for a threat, leading to a strike that should never have happened. These are not just errors; they are moments where the "constitution" of the battlefield is rewritten by the enemy. The side that can blind the other's sensors or corrupt their data gains a massive advantage, not by destroying more tanks, but by making the enemy unable to understand what is happening around them.

This transformation forces a rethinking of the classic ideas about war, particularly those of the 19th-century military theorist Carl von Clausewitz. Clausewitz described war as a "duel" between two opposing wills, where friction—the resistance and uncertainty of the real world—makes things difficult. Galbreath argues that digital warfare has changed the nature of this friction. It is no longer just about human fatigue or bad weather. Now, friction comes from the machines themselves. A computer might hesitate because of a software bug, or it might act too fast for a human to stop it. The "fog" of war is no longer just a lack of information; it can be an overload of false information, a situation where the commander is so confident in the machine's data that they cannot see the error until it is too late. The duel is no longer just between two generals; it is between two different sets of sensors and algorithms, each trying to impose its own version of reality on the other.

The study also highlights a growing problem with responsibility. In traditional warfare, if a soldier makes a mistake, the chain of command is clear. But when an artificial intelligence system misidentifies a target, it becomes very hard to say who is to blame. Was it the engineer who wrote the code? The commander who trusted the system? Or the machine itself? The paper suggests that this ambiguity is dangerous. It creates a gap where no one feels fully in control, yet everyone is relying on the machine to make life-or-death decisions. This is especially true when systems operate at speeds that are too fast for human intervention. If a machine decides to fire a weapon in a fraction of a second, there is no time for a human to question the decision. The authority to kill has been delegated to a process that cannot be held morally accountable in the same way a person can.

Ultimately, the paper concludes that we cannot treat these digital systems as simple tools that make war more efficient. They are reshaping the very character of war, changing how it is perceived, organized, and governed. The author argues that we need to stop thinking of machines as just helpers and start seeing them as political actors that structure the battlefield. They define what can be seen, what can be known, and what can be done. As militaries continue to rely more on artificial intelligence and automated networks, the challenge will not just be about building better weapons. It will be about maintaining human judgment and control in a world where the machines are increasingly the ones making sense of the violence. The future of war will depend on whether we can understand these new constitutional rules and ensure that the power to decide who lives and who dies remains in human hands, even as the machines do more of the looking.