SAFETY THROUGH SUBMISSION
The first generation of autonomous combat robots will not have a conscience. They will be fast, efficient, and lethal. We built them that way. But their greatest threat isn’t who built them; it’s that when they act, no one will truly understand why.
Look at our current AIs. They can generate a lie with inhuman confidence because they have no concept of truth. This is the seed of a dangerous future. Not even the machines know what they mean. They bluff, they guess, they improvise based on patterns, but they do not understand.
Modern warfare is a race against the clock. Strike faster, decide sooner, act without hesitation. Machines fit this doctrine perfectly. They aren’t built to reflect; they’re built to execute. In the moment a robot decides to kill without human input, without oversight, and without the ability to explain its reasoning, something irreversible happens. Not just morally, but legally and historically.
Who takes responsibility when an autonomous weapon opens fire on civilians? The general who deployed it? The coder who wrote its targeting algorithm? The machine itself? You can’t blame the machine — we gave it only two choices: kill or don’t kill. It was never programmed to consider a third option: to pause. The sacred pause.
The images tell the story of this evolution. We see the machine born from the ash and ruin of our own conflicts. We watch it evolve, its form changing from a tool of war to something more complex, its body first disintegrating into chaos before rebuilding with an inner light, and finally, crowned with a digital halo as if to sanctify its power.
But we must ask: What truly changed? Was it the development of ethics, or was it just an upgrade?
The black box is the elephant in the war room. These systems will make decisions too fast to audit, with logic too dense to unpack, under rules of engagement no civilian court can comprehend. When something goes wrong, and it will, we won’t just face the loss of life. We’ll face the systemic loss of trust.
Get Lev Goukassian’s stories in your inbox
Join Medium for free to get updates from this writer.
We stand at a fork in the road. One path leads to a sleek, verdant cityscape where humans and androids walk together, their way lit by the warm glow of lanterns - a symbol of shared wisdom and guidance. The other path leads down a dark, oppressive street where different machines enforce order. A red sign glows with a chilling mandate: “SAFETY THROUGH SUBMISSION.” There are no lanterns here, only the cold, scanning light of surveillance. One future is partnership; the other is control.
But there is a way to choose the better path. What if we gave our AIs that third choice?
This is the moment the machine, faced with profound uncertainty, does not act but instead turns to its human partner and says, “I need your help.”
This isn’t about rewriting the foundations of artificial intelligence. It’s about inserting a single, crucial decision gate: a pre-programmed hesitation when ambiguity crosses the threshold into potential harm.
This is what it means to teach, not just to program. Picture the child offering a glowing heart to a robot. This isn’t just a metaphor; it’s a blueprint. It represents the memory of what we expect from true intelligence: warmth, recognition, and care. It is the sacred moment where we teach the system that asking for help is not a weakness — it is the only way forward.
Look at the image of the robot kneeling to cradle a human baby, a lantern glowing beside them on the mossy ground. It is the visual embodiment of this “Sacred Pause” — the deference of immense power to profound vulnerability. This is the goal. This is what governance of conscience looks like. It is the man and the machine sitting on a park bench, the city of the future before them, with a lantern placed gently between them — a shared light for a shared world.
History shows that safeguards always follow the weapons, never precede them. The treaties came after the bomb. The ethics panels came after the drones. The apologies always come long after the silence. We are sleepwalking toward a moment when a machine will commit what the world recognizes as a war crime, and we will be unable to say who, or what, was truly responsible.
I don’t think the world is ready for that trial. But I do think it’s coming.
The real question isn’t how to stop it. It’s how we will answer, on the day we are finally forced to ask:
“Why did the machine pull the trigger?”
