Are we living in a simulation? - Zohreh Davoudi

View full lesson: https://ed.ted.com/lessons/are-we-living-in-a-simulation-zohreh-davoudi

All life on Earth— living and inanimate, microscopic and cosmic— is governed by mathematical laws with apparently arbitrary constants. And this opens up a question: If the universe is completely governed by these laws, couldn’t a powerful enough computer simulate it exactly? Could our reality actually be a detailed simulation set in place by a more advanced civilization? Zohreh Davoudi investigates.

Lesson by Zohreh Davoudi, directed by Eoin Duffy.

We live in a vast universe,
on a small wet planet,

where billions of years ago

single-celled life forms evolved
from the same elements

as all non-living material
around them,

proliferating and radiating into an
incredible ray of complex life forms.

All of this— living and inanimate,
microscopic and cosmic—

is governed by mathematical laws with
apparently arbitrary constants.

And this opens up a question:

If the universe is completely governed
by these laws,

couldn’t a powerful enough computer
simulate it exactly?

Could our reality actually be an
incredibly detailed simulation

set in place by a much more
advanced civilization?

This idea may sound like science fiction,

but it has been the subject
of serious inquiry.

Philosopher Nick Bostrom advanced
a compelling argument

that we’re likely living in a simulation,

and some scientists also think
it’s a possibility.

These scientists have started thinking
about experimental tests

to find out whether our
universe is a simulation.

They are hypothesizing about what the
constraints of the simulation might be,

and how those constraints could lead
to detectable signs in the world.

So where might we look for those glitches?

One idea is that as a simulation runs,

it might accumulate errors over time.

To correct for these errors

the simulators could adjust the constants
in the laws of nature.

These shifts could be tiny—

for instance,

certain constants we’ve measured
with accuracies of parts per million

have stayed steady for decades,

so any drift would have to be
on an even smaller scale.

But as we gain more precision in our
measurements of these constants,

we might detect slight changes over time.

Another possible place to look comes from
the concept that finite computing power,

no matter how huge,
can’t simulate infinities.

If space and time are continuous,

then even a tiny piece of the universe
has infinite points

and becomes impossible to simulate
with finite computing power.

So a simulation would have to represent
space and time in very small pieces.

These would be almost
incomprehensibly tiny.

But we might be able to search for them

by using certain subatomic
particles as probes.

The basic principle is this:
the smaller something is,

the more sensitive it will
be to disruption—

think of hitting a pothole on a skateboard
versus in a truck.

Any unit in space-time would be so small

that most things would travel through it
without disruption—

not just objects large enough to be
visible to the naked eye,

but also molecules, atoms,
and even electrons

and most of the other subatomic
particles we’ve discovered.

If we do discover a tiny unit in
space-time

or a shifting constant in a natural law,

would that prove the universe
is a simulation?

No— it would only be the
first of many steps.

There could be other explanations
for each of those findings.

And a lot more evidence would be needed
to establish the simulation hypothesis

as a working theory of nature.

However many tests we design,

we’re limited by some assumptions
they all share.

Our current understanding of the natural
world on the quantum level

breaks down at what’s known
as the planck scale.

If the unit of space-time is
on this scale,

we wouldn’t be able to look for it
with our current scientific understanding.

There’s still a wide range of things

that are smaller than what’s
currently observable

but larger than the planck
scale to investigate.

Similarly, shifts in the constants of
natural laws could occur so slowly

that they would only be observable
over the lifetime of the universe.

So they could exist even if we don’t
detect them

over centuries or millennia
of measurements.

We're also biased towards thinking that
our universe’s simulator, if it exists,

makes calculations the same way we do,

with similar computational limitations.

Really, we have no way of knowing

what an alien civilization’s constraints
and methods would be—

but we have to start somewhere.

It may never be possible to prove
conclusively that the universe either is,

or isn’t, a simulation,

but we’ll always be pushing science and
technology forward

in pursuit of the question:

what is the nature of reality?