0:00
Over the straight of Hormuz in the war
0:02
between the United States and Iran, a
0:04
fleet is at work that will never appear
0:06
on the old roles of American naval
0:08
power. Not because it is small or
0:10
unproven, because it moves in silence,
0:13
and there is no one aboard it to name.
0:15
High above the weather, at the very edge
0:17
of the breathable sky, a great winged
0:20
sentinel holds its watch for a day and
0:22
more. Unmanned, sleepless, its gaze
0:24
reaching past the curve of the world. On
0:27
the surface below, a low craft bears
0:29
down on a defended harbor, swift, light,
0:32
and decisive. Along the shipping lanes,
0:35
a lone sail holds its course for months
0:37
on nothing but sun and seaw wind,
0:40
silent, tireless, watching everything
0:42
that passes. And far beneath them all,
0:45
in the black and soundless water, an
0:47
unmanned craft hunts the seabed for days
0:49
on end. No crew, no light, no rest.
0:53
Every one of these is at sea right now
0:55
in the same war over the same narrow
0:58
water. And this is what should give you
1:00
pause. Set them against the way we have
1:02
always reckoned a navy. Hulls, guns,
1:04
crews on the roster, and almost none of
1:06
them register. They are not the carriers
1:08
and the destroyers that rise to mind
1:10
when we speak of American sea power.
1:12
They are quieter and newer and one
1:14
deployment at a time in near silence.
1:17
They are moving from the margins of the
1:18
fleet toward its center. This is the
1:21
force that may yet become the backbone
1:22
of the navy to come. And that is
1:24
precisely why the old measure fails
1:26
here. We have always weighed sea power
1:28
by what can be counted. But a fleet does
1:30
not grow strong by the size of that
1:32
tally. It grows strong by how swiftly it
1:34
can remake itself. And the sea, in a
1:36
real war, never holds its shape for
1:39
long. A force that cannot change with
1:41
it, however vast its count is already
1:43
being left behind. So the question is
1:45
not how many of these machines the Navy
1:47
fields. It is how a fleet built on steel
1:50
and sailors is remaking itself from
1:52
within around vessels that carry no one
1:54
at all and why that transformation
1:56
stands among the deepest turning points
1:58
in the whole history of war at sea. It
2:01
is a turning point the world will only
2:03
recognize long after it is over. On Navy
2:05
Decoded, we would rather not wait that
2:07
long because over the next few minutes,
2:09
this silent fleet is going to show us
2:11
one machine at a time exactly how it
2:19
To see why that change matters, it helps
2:21
to know the fight it grew up in. The war
2:23
between the United States and Iran had
2:25
its center of gravity in one place, the
2:28
straight of Hormuz, the narrow gate
2:30
through which a large share of the
2:31
world's seaborn oil passes every single
2:33
day. When the confrontation over that
2:35
gate escalated, the Navy and its
2:37
partners opened an air campaign,
2:39
Operation Epic Fury, against Iran's air
2:42
defenses, its command nodes, its missile
2:44
and drone sites. Iran did not answer the
2:47
way a weaker navy is expected to answer,
2:49
hull against hull, because it knew how
2:51
that ends. It answered with the cheap
2:53
tools of denial instead. Mines sewn in
2:56
the shipping lane, swarms of fast attack
2:58
boats, anti-ship missiles tucked along
3:01
the coast, one-way attack drones. The
3:04
whole strategy came down to a single
3:05
wager. Make the price of moving through
3:07
that water so high that no one would
3:09
want to pay it. And every tool in it
3:12
leaned on the same assumption. That
3:14
whatever the United States sent into the
3:15
straight would have people inside it.
3:17
People who could be hit or captured or
3:20
made to hesitate. That is the stage. A
3:23
second fact is easy to miss here. The
3:24
crewless war machine at sea is not new.
3:27
It is as old as the aircraft carrier
3:29
itself. Back around the First World War,
3:32
a flying bomb steady by one of Elmer
3:34
Sperry's gyroscopes flew itself out to a
3:36
point a thousand yards away and cut its
3:38
own engine exactly as it had been set to
3:40
do. No pilot, no radio hand on a
3:43
control, just gears and a spinning gyro
3:46
doing the flying. And across the century
3:48
that followed, every war pushed the idea
3:50
one notch further. A small unmanned
3:53
helicopter hunting submarines off the
3:54
decks of Cold War destroyers. a
3:56
jet-powered scout flying the thickest
3:58
air defenses over Vietnam. A spotter
4:01
drone calling fire for battleship guns
4:03
in the Gulf a generation ago. Each step
4:06
was the same move repeated. The fleet
4:08
read a battlefield that had changed and
4:13
So the idea is 100 years old and only
4:16
now is it becoming an entire fleet. Two
4:18
things finally caught up with it. One is
4:20
economics and the other is the machine's
4:22
own intelligence. Drones grew cheap
4:25
enough and autonomous enough to be built
4:27
by the hundred instead of the handful.
4:29
And beneath that sat a sharper
4:31
calculation. The Navy had no wish to
4:33
wager a warship worth billions and the
4:35
sailors aboard it against the cheap
4:37
weapons Iran was putting in the water. A
4:39
mine, a fastboat, a one-way drone, each
4:42
costing a tiny fraction of what it
4:44
threatens. Meeting a disposable threat
4:46
with an irreplaceable ship is a trade no
4:49
fleet wants to make. So the answer took
4:51
shape as a hybrid fleet. crude ships
4:54
backed by hundreds of uncrrewed ones so
4:56
that what is cheap can be answered in
4:58
kind and the hulls that matter are never
5:00
the ones placed on the table. Out in the
5:02
waters around Iran, a unit built for
5:04
exactly this. Task Force 59 had already
5:07
turned the region into a live laboratory
5:09
for the machines. So when Hormuz caught
5:12
fire, the unmanned fleet was not born
5:14
there. It was only put on trial there.
5:17
The easy read of all this runs shallow.
5:19
The obvious point is that drones are
5:20
cheaper and they keep sailors out of
5:22
harm's way. True, useful, and nowhere
5:25
near the whole of it. The deeper point
5:27
is the one the engineer cares about.
5:29
What decides a fleet strength is not any
5:31
single cheap machine, but whether the
5:33
whole force can rebuild itself fast
5:35
enough to keep up with water that will
5:36
not sit still. The drones are just that
5:39
ability made solid. And that ability had
5:41
three very different jobs to do in the
5:43
straight. The fleet had to reach out and
5:45
strike. It had to see everything before
5:47
it was seen. and it had to clear the
5:49
silent road under the surface. And in
5:51
each one, it did not simply acquire a
5:53
tool. It rewrote a piece of itself. The
5:56
sharpest of those tasks comes first, the
6:02
The modern fight put two kinds of
6:04
pressure on the fleet's firepower at
6:06
once. The first is arithmetic. The
6:09
threats coming at it were cheap on
6:10
purpose. Attack drones costing a few
6:13
tens of thousands of dollars. suicide
6:15
boats, mines that run a few thousand a
6:17
piece. And the Navy trained to answer
6:19
with milliondoll precision weapons was
6:22
walking into a losing sum, spending
6:24
fortunes to stop pennies. The second
6:27
pressure is geography. More and more of
6:30
what needed hitting sat in places where
6:32
putting a crew overhead or alongside was
6:35
either too costly or too dangerous to be
6:37
worth it. So, the strike arm did not
6:39
evolve in one direction. It spread out,
6:42
getting cheaper and more expendable on
6:44
one side and reaching into places and
6:46
forms it never had before on the other.
6:49
The clearest sign of the first is a
6:51
weapon the fleet more or less copied
6:52
from the people it was fighting. It is a
6:54
one-way attack drone called Lucas,
6:56
reverse engineered from Iran's own
6:58
Shahed. About 10 ft of airframe, an 8ft
7:02
wingspan, a small piston engine, a reach
7:04
of a few hundred nautical miles, and a
7:07
price tag in the low tens of thousands
7:09
of dollars. It is not sophisticated and
7:11
that is exactly the point. For years,
7:14
the cheap and plentiful side of the
7:15
ledger belonged to the adversary. This
7:18
is the fleet claiming it. An airframe
7:20
cheap enough to launch in numbers, cheap
7:22
enough to lose without flinching. And
7:25
from the opening hours of the air
7:26
campaign, that cheapness went to work
7:28
deep inland. Lucas flying the long
7:31
strike missions against air defense
7:32
sites, command posts, and airfields that
7:35
would once have demanded a crude sordy
7:37
or a far costlier missile to reach. The
7:40
Navy proved it could carry these to sea
7:42
as well, sending one off the deck of a
7:44
latoral combat ship instead of a runway
7:46
ashore, which means the fleet can bring
7:48
a whole magazine of throwaway strike
7:50
drones with it wherever it goes.
7:55
The second pressure reaching where no
7:57
crew could follow shows itself when the
7:59
strike leaves the air altogether and
8:01
comes down onto the water. Now the
8:04
weapon is a small autonomous boat, the
8:06
Seronic Corsair. It crosses the surface
8:09
at better than 35 knots, ranges 1,000
8:11
nautical miles on its own, and steers by
8:14
an onboard navigation fix that needs no
8:16
satellite and no operator. So it holds
8:19
its heading even when every link to the
8:21
outside goes dark. What it proved had
8:24
nothing to do with price. Firepower had
8:26
found a new road. A close-in strike into
8:29
a defended harbor, a run too dangerous
8:31
to ask any crew to make, becomes a
8:34
simple order to give when there is no
8:35
one aboard to lose. A handful of these
8:38
boats drove straight into the ship
8:40
repair yard of a guarded naval base and
8:42
detonated themselves against it. It was
8:44
the first time American forces sent
8:46
surface robots not to watch, but to
8:49
strike. The attack had taken a shape it
8:51
had never had before. And at the top of
8:53
this layer sits a machine that already
8:55
carries both halves of the story, a
8:58
weapon and an eye in the same airframe.
9:00
The MQ9 family is a hunter and a watcher
9:03
at once. It carries the missiles, but it
9:06
also carries the camera that finds the
9:07
target in the first place, and it does
9:09
both with no pilot in a cockpit. Over
9:12
Hormuz, though, that same aircraft
9:14
showed the fleet that the water had
9:16
shifted under it again. Flying low and
9:18
slow enough to study the surface put it
9:20
right inside the reach of defenses that
9:22
had learned its habits. And the fleet
9:24
lost these drones fast. Something close
9:27
to a quarter of the force over the
9:28
straight. That is not a defeat so much
9:31
as a message. When keeping an eye on the
9:33
water costs a quarter of your aircraft,
9:35
the job of seeing has come apart from
9:37
the job of hitting, and it needs an
9:39
answer of its own. Which is the thing
9:41
worth carrying forward? Not the missile
9:43
the MQ9 slings, but the eye it flies
9:46
for. Holding that eye over the straight
9:48
the old way costs a price the fleet
9:50
cannot keep paying. So the next move had
9:53
nothing to do with hitting harder. It
9:55
was learning to see in a way that
9:57
survives. And that is where the fleet's
9:59
eyes begin to pull away from its fist.
10:03
In every battle, on every battlefield,
10:06
one thing weighs more than steel or
10:08
numbers. Information. Knowing the shape
10:11
of the fight before it breaks over you.
10:14
And the thing that feeds that knowledge
10:15
to the fleet is its eyes, the machines
10:18
that watch. That knowledge is never only
10:21
a shield. The same picture that shows
10:23
where the mines lie and where the fast
10:25
boats are gathering is the picture that
10:27
tells the fleet's weapons where to
10:29
strike. Information cuts both ways. It
10:32
keeps the fleet alive, and it puts the
10:33
target in the crosshairs. So, the
10:36
fleet's eyes had to evolve along one
10:38
unforgiving line, to see farther, to see
10:41
and survive, and to keep watching even
10:43
when the signals went dark. The eye that
10:45
fell over Hormuz had already taught the
10:47
lesson, "Fly low enough to read the
10:49
water, and you fly into range of
10:51
everything built to bring you down."
10:53
Even the Navy's maritime version of that
10:55
drone, the Sea Guardian, fitted with
10:57
sonoys to hear submarines below, as well
10:59
as watch the surface, lived in that same
11:01
exposed band of sky. So survival forced
11:04
the eye to evolve in two opposite
11:06
directions at once. Climb until nothing
11:08
can reach you or grow so cheap that
11:11
losing you costs nothing at all. Those
11:13
two directions became the next two
11:15
machines. Go up and you arrive at the
11:17
Triton, a drone with a wingspan wider
11:20
than an airliners, flying its watch
11:22
above 60,000 ft. Higher than the
11:24
weather, higher than almost anything
11:26
built to reach it. It holds there for
11:28
more than a day on a single tank,
11:30
sweeping a full circle of ocean with one
11:35
The physics of why it climbs so high is
11:37
worth one clean sentence because it is
11:40
the quiet key to the whole layer. The
11:42
Earth curves, so every radar down on the
11:44
surface is boxed in by its own horizon.
11:46
And the higher you rise, the farther
11:48
that horizon runs. from 60,000 ft. This
11:51
one aircraft sees more sea in a single
11:54
turn than a squadron of ships strung
11:56
across the water. It is expensive and it
11:58
is rare and losing one near the straight
12:00
costs the fleet dearly. But altitude
12:03
buys the fleet something no crude hull
12:05
can. A look that reaches past the curve
12:07
of the planet and holds there for a day
12:09
at a time. And it does not only look, it
12:12
passes that picture back to the whole
12:13
force in real time. A single node in the
12:16
sky knitting scattered ships into one
12:18
set of eyes. Go the other way, all the
12:20
way down the scale, and you reach a
12:22
machine an adversary cannot make
12:24
expensive, no matter what it tries. It
12:26
is a sail drone, a sailboat with no
12:28
crew, driven by wind and sun, that holds
12:31
its station for months without
12:32
refueling, finds its own way even when
12:35
its satellite fix is jammed and cost
12:37
almost nothing to keep at sea. It
12:39
carries no weapon and it is in no hurry.
12:41
Its whole talent is that it never
12:43
blinks. And there is a quiet proof of
12:45
how far this pushes the idea. When an
12:47
adversary went out and physically seized
12:49
one of these sailboats, what it hauled
12:51
aboard was a camera that runs on wind.
12:53
No crew to parade, no secrets worth the
12:56
taking, a hull worth almost nothing. An
12:59
eye you cannot wait out, cannot make
13:01
costly, and cannot even turn into a
13:03
prize by capturing it. So the eyes
13:05
evolved, reaching farther, harder to
13:07
bring down, impossible to tire out. But
13:10
seeing the minefield and the boat swarms
13:12
is only half the work. Something still
13:14
has to go into that minefield first.
13:16
feel out its edges and open a lane. And
13:18
that work happens in the one place no
13:20
camera above the surface can follow
13:22
underneath it in the quiet where the
13:24
whole fleet supply lines are kept alive.
13:28
The strike hits and the eyes see, but
13:31
the straight does not open until the
13:32
road beneath the surface is clear. And
13:35
that road runs through the one part of
13:37
the fight the fleet had never been able
13:38
to hold for long. the deep quiet water
13:42
where a divers's time is counted in
13:44
minutes and a crude submarine cannot
13:46
afford to loiter. It is also where the
13:48
crulest of the cheap weapons waits, the
13:51
mine. To keep up, the fleet had to grow
13:53
a presence in a place it had always been
13:55
half blind. The quiet answer is a
13:57
machine that draws no headlines, an
13:59
uncrrewed submarine the Navy calls the
14:02
dive LD. It is about 19 ft long, a few
14:05
tons of pressure hardened hull, built to
14:07
work for 10 days at a stretch with no
14:09
one tending it and no support ship
14:11
holding its hand. And much of its body
14:13
is simply printed layer by layer so it
14:15
can be built fast and reconfigured
14:17
cheaply. Its jobs are the unglamorous
14:19
ones that make everything else possible.
14:21
Charting the seabed, relaying
14:23
communications through water that
14:24
swallows every signal, hunting for what
14:26
should not be there. And the reason it
14:28
cannot be swapped back for a person is
14:30
plain physics. Pressure in the deep
14:32
crushes anything soft. A diver cannot go
14:35
there at all. And a crude submarine
14:37
cannot sit over one patch of water for
14:39
10 unbroken days. But a few tons of
14:42
quiet machine with nothing inside to
14:43
protect can. It asks for almost no
14:46
logistics train behind it. It goes out,
14:49
does the work, and comes home. The fleet
14:51
has grown a sense where it used to be
14:55
Nowhere is that clearer than in the
14:57
minefield itself. The straight was
14:59
seated with more than a hundred mines,
15:00
and a single mine, a few thousand
15:02
dollars of it, can [ __ ] a crude
15:04
warship worth billions. That is the
15:06
asymmetric problem in its purest form.
15:09
The fleet's answer was to send the
15:10
machines in first. Robots like the knife
15:12
fish that sweep the bottom with sonar,
15:14
sorting the real threats from the
15:16
clutter, even finding the mines buried
15:18
in the silt, while the ships that carry
15:19
them wait safely outside the field. This
15:22
is the moment where no one aboard stops
15:24
being a convenience and becomes a whole
15:26
new capability. The fleet can now put a
15:28
machine into water where every lane used
15:30
to be a gamble weighed against the ship
15:32
and everyone riding it and send in the
15:34
machine instead of the crew. The same
15:36
quiet logic runs all along the fleet's
15:38
floor. Uncrewed boats stand in as
15:40
communications relays. In the air, the
15:43
idea of a drone tanker stretches every
15:45
strike fighter's reach. Logistics is the
15:47
front no one televises, and without it,
15:49
the strike and the eyes are worth
15:51
nothing. Three layers now have each been
15:53
rewritten. the Strike, the Eyes, and the
15:56
Road below. But the thing that turns
15:58
three separate layers into a single
15:59
fleet lives in none of them. It lives in
16:02
the fact that not one of these machines
16:03
will stay in the lane it was built for.
16:09
Put it together, a fleet that strikes
16:10
cheap, watches without tiring, and
16:13
reaches into water it once feared. And
16:15
the deepest sign of how far it has
16:16
changed is not on any one machine's spec
16:19
sheet. It is in what the machine started
16:21
doing to each other's jobs.
16:24
Take that windpowered scout, the one
16:26
whose whole talent was never blinking. A
16:28
larger cousin of it was fitted with a
16:30
missile launcher and in an exercise
16:32
fired two guided rounds into a fastboat
16:34
target. And the shot was aimed not by
16:36
anything on the sailboat, but by the
16:38
sensors of a carrier strike group tens
16:40
of nautical miles away, and with it the
16:43
crew aboard that ship, who held the
16:45
decision to fire. The machine carried
16:47
the weapon, not the authority to use it.
16:50
The watcher had become the shooter and
16:52
the round was guided by a ship sitting
16:54
over the horizon. Sensor and weapon
16:56
pulled apart and rejoined across a
16:57
network in a single frame. And it runs
17:00
the other way, too. The same class of
17:02
attack boat that ran into that defended
17:04
harbor is the same kind of hull that
17:06
once pulled two aviators out of the
17:07
water off Hormuz. The machine that
17:10
strikes is the machine that rescues. Not
17:13
one of these things will agree to be a
17:15
single thing. And that refusal is the
17:17
shape of a force that has learned to
17:18
swap roles faster than the fight can
17:20
change the question. This is where the
17:22
change stops being a run of upgrades and
17:24
becomes a change in kind. An old fleet
17:27
is a collection of hulls, each welded to
17:30
one job. The fleet taking shape is a
17:32
pool of interchangeable
17:34
nodes, sensor, shooter, tanker, mine
17:37
hunter, lifeboat, where the role flows
17:39
from one machine to the next as the
17:41
moment demands, and where losing a node
17:43
just makes the network route around the
17:45
gap. It is a force designed to treat
17:47
change as a feature instead of suffering
17:49
it as a shock. That is why the arrival
17:51
of these machines in the United States
17:53
Navy is not some sudden out of nowhere
17:55
event. It is one of the deepest
17:57
strategic and technological turning
17:59
points in the whole history of war at
18:00
sea. Not a gadget bolted onto an old
18:03
fleet, but a fleet that has learned to
18:04
reshape itself faster than the fight can
18:07
reshape the problem in front of it.
18:09
Hormuz did not create that. Hormuz only
18:11
proved it. And the answer to how a fleet
18:13
actually gets stronger lives in none of
18:16
these machines by itself, but in the
18:18
refusal to ever finish becoming
18:25
So what is happening here goes well
18:27
beyond a navy buying new tools. A force
18:30
is learning to treat change itself as a
18:32
capability to reshape itself faster than
18:34
the fight can throw new problems at it.
18:36
That is why the rise of these machines
18:38
is no passing fashion. It is a century
18:40
of a fleet teaching itself to change,
18:43
finally cast in steel and code, shifting
18:45
the ground beneath sea power itself,
18:47
away from the strength of what a navy
18:49
owns and can count toward the strength
18:51
of what it can become. The evolution is
18:53
nowhere near finished. Drones still fall
18:56
by the dozen when the defenses learn
18:58
their patterns. A spoofed signal can
19:00
still turn an autonomous hull against
19:01
its owner. The law of the sea has not
19:04
even decided whether a ship with no crew
19:05
is a warship at all. But for a fleet
19:08
built to adapt, none of that is failure.
19:11
Each is only the next problem to evolve
19:13
against. Which leaves a harder question
19:14
than the one we began with. When
19:16
advantage at sea is no longer measured
19:18
by what a fleet holds, but by how fast
19:20
it can remake itself. When the thing you
19:22
cannot destroy is not a ship, but a mesh
19:25
that rroots around every loss, then the
19:27
contest between navies stops being about
19:29
who can build the most and becomes about
19:31
who can learn the fastest. That race has
19:34
no finish line because the sea never
19:36
stops changing. The decade ahead will
19:39
not belong to the largest fleet. It will
19:41
belong to the one that can change in