THE FUTURE
NEEDS A JERK.

“accelerate” is an instruction with a missing object. accelerate the cure, the surveillance, the extraction, the recovery, the concentration of ownership: the verb approves all of them equally. if your philosophy cannot distinguish between those destinations without importing values from somewhere outside the speed itself, then speed was never your philosophy. it was the part that let you avoid having one.

showing that a system can do more establishes a capability. it does not establish that its deployment is wise, that its owners deserve political authority, or that the people who bear its costs have agreed to the experiment. each of those claims requires its own argument. sliding from the first to the others while the audience is still impressed is how a technical achievement becomes a sales pitch for power.

the good future has actual conditions. people can eat, think, refuse, create, love, and participate in decisions that govern them. count the hours a technology gives back. count the dependencies it creates. count who can leave and who has to keep paying. if the curve is going up while those conditions deteriorate, have the intellectual decency to say what the curve is measuring.

we want breakthroughs. we want the extraordinary expansion of what human beings can do. that is precisely why “more” is such a pathetic final answer to the question of what we should do with it. a civilization capable of inventing these machines should be capable of specifying a purpose beyond feeding the next machine.

you learned the vocabulary. you can explain the incentives, diagnose the institutions, identify the propaganda, and predict the failure. excellent. now point to the obligation this knowledge has imposed on you. what did you stop doing? what did you risk? what did you build that was harder than announcing that you understood?

there is an entire kind of self-satisfaction available to the person who can describe a catastrophe in sophisticated language while remaining comfortably adapted to it. the diagnosis becomes a credential. the credential becomes a personality. eventually the person needs the world to remain broken because being the person who sees through it has become their most reliable source of importance.

and certainty makes the trap worse. once “awake” becomes something you are rather than something you keep doing, disagreement starts to feel like an insult to your rank. you demand that other people question their conditioning while treating your own conclusions as the end of the investigation. congratulations: you have reproduced the intellectual obedience you claimed to escape, with yourself installed as the authority.

an effective jerk has to survive the same treatment they administer. bring the objection that could change your mind. name the evidence that would defeat your favorite explanation. allow the person with the less impressive vocabulary to notice the thing you missed. the capacity to be corrected is the part of intelligence your ego will try hardest to counterfeit.

take that sentence seriously and an uncomfortable distinction appears. a system can become more capable at the same time that the people dependent on it lose the ability to act independently. it can improve its performance while your alternatives disappear. the intelligence demonstrated in the product and the intelligence available to the person using it can move in opposite directions.

imagine the assistant that makes your work effortless, then becomes the only place your work can exist. the service that gives you an audience you cannot reach without its permission. the infrastructure that becomes indispensable before its owner changes the price. the initial convenience is real. so is the dependence. the first does not cancel the second.

ask what you can take with you. ask whether another tool can read your work. ask whether you retain the skills, rights, relationships, and resources required to leave. an exit that exists only in a legal document, at a cost no ordinary person can survive, is an exit designed to remain unused.

the standard is demanding because it should be. build systems that make people more capable beyond the boundaries of your product. let them understand, modify, migrate, and refuse. if the business only works while those possibilities remain unavailable, stop presenting its growth as a straightforward increase in human freedom.

imagine a system that reconstructs the voice of someone you lost, learns the details you preserved, and responds with the gestures of a relationship you would give almost anything to recover. now imagine a company asking you to treat that resemblance as evidence that the person has returned. the emotional force of the encounter would make the question harder to ask at exactly the moment it became most necessary.

what has actually been reconstructed? a pattern of expression, a collection of memories, a conscious subject, a continuation of the same life? these are different claims. moving between them on the strength of a grieving person’s relief would be an extraordinary abuse of trust. making a representation more convincing does not, by itself, settle which claim is true.

the same discipline applies to the grander promises: artificial gods, simulated universes, technological transcendence. take the possibilities seriously enough to distinguish a conjecture from a conclusion. a simulation argument rests on assumptions; it does not give you a discovered creator or resolve the nature of consciousness. [3] confidence delivered at scale cannot perform the missing philosophical work.

there is room here for faith, reverence, and the recognition that reality exceeds what we understand. there is also a profound difference between approaching that mystery with humility and selling access to a machine as if you have acquired the rights to it. if an engineered experience becomes indistinguishable from revelation, the identity and incentives of its engineer become more important, not less.