Draw the draft

They might actually be used.… They stopped calling it “war” in official briefings. By 2026, the planners in Tokyo, Washington, and Brussels all used the same phrase: “time-compressed conflict.” The assumption was simple—wars had to end quickly or spiral out of control. Supply chains snapped in weeks. Satellites were blinded in hours. Financial systems destabilized in minutes. No one had time for total war anymore. Colonel Sato watched the simulation unfold on a wall of screens. Hypersonic missiles crossed the Taiwan Strait in under ten minutes. Autonomous drones hunted radar signatures. AI systems flagged targets faster than humans could authorize strikes. And above it all—untouched, unlaunched—sat the nuclear arsenals. They were never part of the simulation. Not really. The Shadow That Isn’t Used Sato had studied the Ukraine war years earlier. Not the battles—the absence. Russia never used nuclear weapons. Neither did ...

Or who, exactly, the enemy was.… The maps were wrong. They still showed NATO as a single, clean shape—thirty-two countries shaded in calm blue, as if unity were a geographic fact. But in the operations room beneath Brussels, the officers had stopped looking at maps months ago. They watched bandwidth. “Split traffic again,” said Colonel Ionescu, pointing at the wall of live feeds. “U.S. channels are throttling European command relays.” No one looked surprised. Since the Greenland crisis of 2025–2026, the alliance had begun to behave less like a bloc and more like a negotiation that never ended. The Americans called it “burden correction.” The Europeans called it something else, quietly, in their own languages. A technician pulled up Arctic satellite imagery—synthetic aperture radar overlays, the same systems once used to track Russian armor in Ukraine. Now they were watching each other. “Pituffik base expansion confirmed,” she said. “New runway segments. Inc...

Fingers on the trigger.… The satellite passed overhead at 02:17. From 500 kilometers above the earth, its synthetic aperture radar swept across the valley—through clouds, through darkness, through camouflage nets stretched like skin over armored vehicles. The data streamed down instantly: heat signatures, vehicle outlines, probability models. Within seconds, analysts knew there were twelve armored units, three artillery positions, and a logistics convoy staged near the river. It was perfect reconnaissance—broad, deep, untouchable. And completely insufficient. Because by 02:23, everything had moved. Captain Arai stared at the updated feed. The satellite pass had already gone stale. Orbital mechanics didn’t care about urgency—there would be no second look for another ninety minutes. “Deploy the drones.” He didn’t say reconnaissance drones. Nobody used that term anymore. ⸻ The quadcopters lifted silently from the hillside....

And trust, once optimized away, does not return.… The city still voted. Ballots were cast, counted, verified—digitally signed, redundantly stored, audited by three independent systems. On paper, everything was intact. International observers praised the process. The turnout had even increased this year. And yet, no one believed the government had authority. Not anymore. Aya Nakamura worked inside the Ministry of Civic Coordination, a department created after the “Optimization Reforms” of the early 2020s. Officially, its role was simple: improve efficiency in welfare distribution, infrastructure planning, and emergency response using predictive algorithms. Unofficially, it decided who got what—and when. Housing permits. Medical prioritization. Disaster relief routing. Business subsidies. Every benefit flowed through a system called CIVIS, a machine-learning infrastructure that processed millions of data points in real time. It did not command. It did not thre...