Explainer
Trillions of Unborn People Rely on One Century
Why total future lives dwarf everything happening right now.
In 2003, philosopher Nick Bostrom sat down to calculate the moral cost of human extinction, not in tears or historical grief, but in lost inventory. He estimated that if human descendants eventually settle our local virgo supercluster, running digital minds on solar infrastructure, the universe could comfortably support roughly 10^38 human-equivalent lives per century. Run that for a few billion years, and you realize that every standard moral emergency we argue about today—famine, war, malaria—is, by sheer arithmetic, an infinitesimal rounding error on a ledger of cosmic magnitude.
Key point
If future people have equal moral weight, current human suffering becomes a mathematical rounding error.
The moral stance
Total utilitarianism rejects distance in time: a person living ten thousand years from now has the exact same moral claim to exist and flourish as your next-door neighbor.
The technological multiplier
The transition to recursive self-improvement and artificial superintelligence acts as an irreversible pivot point, permanently determining whether that future inventory expands or gets wiped out entirely.
The Kicker
When the denominator is astronomical, reducing extinction risk by even 0.0000001% outweighs solving any crisis facing people alive today.
This logic transforms the century of machine intelligence into what Toby Ord terms the hinge of history. In normal times, human decisions govern five-year budgets or fifty-year treaties; right now, an engineering lab in San Francisco tweaking loss functions might permanently close or unlock an ocean of quadrillions of potential minds. If an unaligned ASI emerges and sanitizes the biosphere next Tuesday, the real tragedy in this framework is not the death of eight billion existing bipeds, but the retroactive cancellation of an unimaginable multitude who will never get to exist.
It sounds noble until you watch the ledger operate on live subjects. Treat unborn minds as real stakeholders, and you inevitably reach the uncomfortable conclusion: spending resources to heal an existing sick child is practically criminal negligence if that same dollar could purchase a fractional basis point of AI alignment security.
Analogy · ETHICS & SCALE
This Concept
Existential risk mitigation
- →Trillions of hypothetical future minds
- →Single brittle century of AI transition
- →Current living generation
- →Permanent loss of cosmic potential
Is Like
Carrying an heirloom vase
- Generations of family inheritance
- One slippery patch of ice on the stairs
- The courier currently holding the box
- Dropping the ceramic into powder
Where it breaks down
A dropped vase only ruins an object, whereas failing the AI transition cancels the existence of observers who cannot protest their own erasure.
Here lies the turn: the math works perfectly, but the constituency is entirely silent. An ideology that demands we prioritize ghost populations across deep time leaves living humans as mere scaffolding, required only to build the machine safely and exit without breaking the glass. When you weight potential lives identically to existing ones, ethics stops being a conversation about compassion and becomes an extraction problem: how much present agony can we justify to guarantee an astronomical spreadsheet?
Tensions behind the calculation
ai
Thought Experiment
ai
Concept
ai
Satire Article
That tension is where the philosophy currently fractures. If superintelligence is the sole gateway to this cosmic inventory, does guarding against speculative catastrophe give us license to ignore every injustice happening in the room right now?
Choose