Please join my mailing list here 👉 https://briankeating.com/yt to win a meteorite 💥 What if the equations that describe every swirl of water, every gust of wind, and every turbulent storm contain a hidden failure mode?
In this episode, I explore the mystery at the core of the Navier–Stokes equations: in three dimensions, we still do not know whether smooth initial conditions stay smooth forever, or whether the mathematics can blow up into a singularity in finite time. That question is so fundamental that it became one of the Clay Millennium Prize Problems, with a one-million-dollar reward for a definitive proof.
We break down what makes the equations hard: the nonlinear self-advection term, where the fluid effectively carries itself. This self-referential loop is the engine of turbulence and the reason tiny perturbations can amplify into chaotic motion.
Then we follow recent guest Terence Tao’s strategy for attacking the problem: tracking how energy cascades from large swirls down to smaller and smaller eddies, and searching for invariant quantities (critical norms) that could mathematically prevent runaway concentration.
Finally, we enter the modern twist: AI as a partner in discovery, scanning massive turbulence data for hidden patterns and helping build better, simplified models that can guide human proofs.
If the equations fail, it is not just a technicality. It means our map of fluid reality is incomplete.
Chapters 00:00 The “math glitch” hiding in everyday fluid flow 01:13 From Newton’s F=ma to Navier and Stokes 01:54 Why Navier–Stokes powers planes, weather, and simulations 03:09 What the terms mean: velocity, pressure gradient, viscosity 03:43 The troublemaker: nonlinearity (u·∇u) and why chaos emerges 04:21 Laminar vs turbulence: two regimes under one equation 05:00 Engineering turbulence versus solving it exactly 05:24 The one million dollar challenge: what must be proved 05:41 Two outcomes: global smoothness or finite-time blowup 06:34 What a singularity means: infinite velocity or pressure 07:16 Why it matters even if real water never goes infinite 07:58 Tao’s approach: a nearby “toy universe” with provable blowup 08:30 The warning shot: the real equations may sit on a knife-edge 09:03 Energy cascade: how turbulence moves energy to small scales 09:51 Invariants and critical norms: the mathematical seatbelts 10:38 Enter AI: pattern hunting in high-dimensional turbulence 11:13 How AI helps: invariants, toy models, and proof guidance 12:12 The precipice: cracking the code of chaos 13:04 Outro
📚 Get my books: Think Like a Nobel Prize Winner, with productivity tips from 9 Nobel Prize winners: https://a.co/d/03ezQFu
Focus Like a Nobel Prize Winner, with life-changing interviews with 9 Nobel Prizewinners: https://a.co/d/hi50U9U My tell-all cosmic memoir Losing the Nobel Prize: http://amzn.to/2sa5UpA
The first-ever audiobook from Galileo: Dialogue Concerning the Two Chief World Systems: Ptolemaic and Copernican https://a.co/d/iZPi9Un
This video contains commentary and criticism under Fair Use (17 U.S.C. §107). All third-party clips, articles, and documents are used for educational and critical purposes.
Look closely at the world around you.
Whether you're observing the majestic
spiral arms of distant galaxies or
watching the milk swirl into your
morning coffee, [music]
you are witnessing the same fundamental
phenomenon, fluid flow. It's beautiful.
It's mesmerizing. And it seems to be
consistent until we look deeper.
We believe that a single set of rules
governs all of this motion. From the
largest cosmic scales down to the
smallest eddies in a stream, but there
is a haunting question that keeps
physicists and mathematicians like
Terrence Tao awake at night. What if
those rules can break? What if the
mathematical code that describes this
motion contains a hidden flaw?
Today, we're embarking on a journey into
the hidden heart of complexity, where we
ask if our current understanding of the
physical universe is complete or if
there's a glitch waiting to be found in
the very nature of how fluid and other
particles move.
To understand the glitch, we first have
to understand the rules. For that, we
need to travel back 200 years to the
1800s. Two brilliant minds, Claude Louie
Navier and George Stokes, penned what we
now call the master equations for fluid
flow.
Starting with Isaac Newton's deceptively
famous and simple second law, force
equals mass time acceleration, which
works perfectly for solid blocks sliding
down ramps, say, but they adapted it for
a continuous substance. This was a
massive leap. They created a
mathematical recipe to describe
substances that don't hold their shape
like water or air. These equations,
later known as the Navier Stokes
equations, are the bedrock of modern
physics. They give us instructions for
how fluids move, how fluid flows, how
ocean waves swirl, and describe a lot of
video game and AI generated physics
demonstrations. And they provide the
basis for modern simulations, including
their incarnations in video game design.
They're the very reason we can design
airplanes and predict weather. But
writing down the equations is one thing.
Solving them is something totally
different. This mathematical sentence is
the source code for classical physics
regarding fluids. It looks intimidating,
but it actually tells you something
dynamical. It represents a precise
balance of forces. On the left side, we
have the acceleration of the fluid, how
its motion is changing from one moment
to the next. On the right, we have the
forces pushing and pulling on that
fluid. F= ma. It's an equation that
updates our reality and our
understanding millisecond by
millisecond. It tells the water in a
pipe or the wind over a wing exactly
what to do next. But there's a catch.
While the equation looks compact, it
hides infinite complexity.
It's what's called a differential
equation.
And such equations must hold true at
every single point in 3D space all at
once. And hidden within these symbols is
a specific mechanism that makes the math
incredibly difficult to untangle. Let's
break the equation down into its
characters to see where the drama comes
from. We have the velocity which tracks
changes over time. We have the pressure
gradient which dictates how fluid moves
from high pressure to low pressure. We
have the viscosity term which represents
internal friction. Think of it as the
stickiness of honey that tries to smooth
things out. But then there's the
troublemaker, the term in the middle, U.
Dell * U. This represents nonlinearity.
U* U. U squared. In simple terms, this
is the fluid carrying itself around. It
means that the flow of energy depends on
the flow itself.
It seems like selfdealing. But this
self-reerential loop is where chaos is
born. It turns simple inputs into wildly
unpredictable and complex outputs.
Whenever you see nonlinearity, these
types of behaviors arise when the other
terms try to keep things orderly and
smooth. This nonlinear term is
constantly trying to twist the system
into unpredictable knots, whirls,
eddies, vortices. Because of that
troublemaker term, these equations
govern two completely different worlds.
On one side, we have laminer flow, the
predictable world. This is smooth,
gentle, and soluble. We can calculate
exactly what a slowmoving river will do.
But turn up the speed and you enter the
second world, turbulence. This is the
realm of chaos.
In turbulence, tiny changes at the
beginning magnify. They don't fade away.
They get amplified.
A butterfly flapping its wings causing a
tornado is not just a metaphor. It's a
mathematical possibility here. We use
these equations every day to engineer
technology
to design airplanes and wind turbines
and even to mimic reality in video games
so you can get that extra millisecond of
frag speed. But the frightening truth is
that for turbulent flow, we can only
approximate the chaos. We cannot solve
it exactly. We literally do not know if
the math holds together or whether it
breaks down when the water gets rough.
This uncertainty is so profound that the
Clay Mathematics Institute has named one
of the seven Millennium Prize problems.
They have placed a $1 million bounty on
the answer. The challenge is not to
calculate a specific flow, but to prove
something fundamental about the
equations themselves in three
dimensions. The challenge asks for a
proof of one of the two following
scenarios. Scenario A, do solutions
always exist and say smooth and well
behaved forever? Or scenario B, can you
show a specific example where a smooth
start leads to a disaster? The
millennium problems ask for a
mathematical guarantee.
Either the universe is safe from runaway
chaos and the equations will work
forever or there's a trapdoor in reality
where the physics literally falls apart.
What does it mean for physics to fall
apart, though? Is it just a catchphrase
that mathematicians use to make
something sound more breathlessly
exciting than it actually is? No.
Physicists and mathematicians call
something a singularity or a blowup.
Whenever a following phenomenon takes
place, imagine a fluid flowing smoothly.
Then purely through its own internal
mechanics, it begins to twist and
tighten. It forms a knot of energy that
gets smaller and smaller, spinning
faster and faster.
A blowup or singularity is the
theoretical moment where the velocity or
pressure at that point becomes infinite.
And the terrifying part is that the math
suggests this could happen in a finite
amount of time. It's like a camera
zooming in on a whirlpool and before you
can stop it, the speed hits infinity and
the laws of physics break. If this can
happen, it means the equations describe
a universe that will eventually destroy
itself. at least the fluid mechanical
part of itself. You may ask, why should
I care about a math glitch? Real water
doesn't explode like this. And you'd be
right. And that's what led Terrence Town
on his adventure to try to solve this
problem. Real water does not possess
infinite velocity. But that is exactly
why this matters. If the Navier Stokes
equations predict a blowup, but nature
never actually demonstrates one, it
means our equations are fundamental
source code for classical world are
flawed or perhaps incomplete. It could
be a crisis for physics if this were
true. It would imply that the tool we
have used for two centuries to build our
civilization or at least the technology
in it is wrong at a very very deep and
fundamental level. It tells us that our
map does not match the territory.
Solving this isn't just about the money.
Obviously, it's about verifying that we
actually understand how the physical
world is constructed.
For decades, no one could prove it
either way. Then in 2014, Terry Tao got
on the scene. He couldn't initially
prove a blowup in our real equation. So,
he tried something clever. He built a
toy universe. He created a mathematical
model with equations that were slightly
modified, very similar to ours, but
distinct. In this parallel mathematical
universe, Terry proved that a finite
time blow up or singularity is possible.
He showed that fluid in this toy model
could indeed crush itself into a
singularity. This was a warning shot. It
wasn't real life. It wasn't our real
universe, but it showed us that the real
Navigator Stokes equations are likely
sitting right on the cusp of a
singularity of a blowup.
We are balancing on the knife edge
between stability and chaos. To solve
the real problem, Tao developed a road
map focused on what's called the energy
cascade phenomenon. In turbulence,
energy moves. It transfers from large
scales like big swirls in a river down
to smaller scales and then even smaller
scales.
Eventually, we end up in tiny almost
microscopic eddies. Usually, the
viscosity kills the energy at the
bottom. But the danger zone is if this
cascade happens too fast. If the energy
concentrates into an infinitely small
point faster than friction can dissipate
it, you get a blow up, a singularity.
Ta's program is essentially an attempt
to track the energy, to bookkeep it, to
make sure that we can prove
mathematically that the energy can never
pile up fast enough to break the system.
He needed to tame the cascade. The
strategy to tame the cascade relies on
finding invariance.
These are things that due to symmetry
never change.
Think of conservation of energy in the
chaotic mess of a storm. Are there
quantities that refuse to change? Are
there symmetries we can exploit? These
are called critical norms. If you can
find a quantity that remains invariant
under scaling, meaning it doesn't get
bigger even as the swirls get smaller
and smaller, you have a leash to tame
the chaos. The mathematical goal is to
prove that these critical quantities
stay bounded, finite. If they have a
limit, then the velocity can never go to
infinity and the system can never really
blow up. We are hunting for the
mathematical seat belts that keep the
universe from falling out of the car,
but finding them has proven quite
difficult for human intuition alone.
This is where the story enters the
modern age. The complexity of 3D
turbulence might be beyond what human
brains can visualize on their own. So, a
new partner has joined the hunt. AI,
artificial intelligence. We aren't just
using computers to run simulations and
make better video games. We're using
deep learning and AI reasoning models to
think alongside mathematicians like
Terry. These AI models can navigate
highdimensional spaces that leave human
brains baffled, even brains as bright as
Terry Ta. They serve as a partner for
mathematicians and look for the breakage
points and blowups and singularities in
the math. It's a fusion of rigorous
human logic and the unmatched pattern
recognition prowess of machines.
How does this partnership actually work
though? The AI follows Terry's road map
in two specific ways. First, it acts as
a pattern hunter. It analyzes massive
data sets of turbulent simulations to
find those hidden invariants, the
statistical norms, the laws that humans
simply can't see in the bulk amidst the
noise. Secondly, the AI that they have
developed helps build better toy models,
more accurate to help us reverse the
engineering data to propose new
simplified equations for mathematicians
and physicists to study. In this way, AI
is helping our intuition by bridging the
gap between the unsolvable real world
and the solvable but incomplete toy
models. It's checking the math, probing
for weaknesses, and guiding the human
researchers towards the final proof.
We're now standing at the precipice. The
Navier Stokes equation solutions,
unsolved for nearly two centuries in
their full glory, is finally within
grasp. It's coming from a perfect storm,
a pairing of Terrence Ta's brilliant
human road map guiding the immense
computational power of AI. Together,
Terry and his collaborators are doing
more than just hunting to solve a $1
million prize. They are decoding the
fundamental rules of our universe.
Whether they prove that the universe is
safe and smooth or that it contains a
hidden singularity, the answer will
change physics forever.
We're about to learn if the code of
chaos can finally be cracked. And stay
tuned for more episodes about these
vexing problems and what they mean for
both physics, mathematics, and even
video games.
And click here to watch my two
interviews with Terrence Ta. one walk
and talk on the future of science and
the other a sitdown interview in his
office at UCLA about his work about all
of his explorations in mathematics.
Don't forget to like and comment and
subscribe.