There's a whole new economy, where things are made and assembled in space.

From huge satellites and spacecraft in orbit, to entire cities on the Moon and Mars.

We are treading new ground here.

So what advances in technology could make building and manufacturing in space a reality?

The answer could be robots, guided by AI.

I think a lot of paradigms will change the moment we start manufacturing in space.

Ed Mayer is co-founder of a company with ambitious plans to build a factory of the future in space.

AI and robotics are going to be a big part of that.

The ability to do this kind of building in space is decades away.

But Machina Labs is drawing inspiration from the manufacturing process that it's currently using here on Earth.

We call these robotic cells Robocraftsman.

The company specialises in a new way of shaping sheet metal, called roboforming.

Think of two robotic arms on two sides of a flat sheet of metal, working together like fingers of the potter, to deform a sheet of metal into very complex shapes.

They can pick up different tools, apply it differently to the material, and replicate all kinds of different processes.

The hope is that one day, a new generation of robots like these will be able to assemble and construct things in space.

Welcome to our second facility.

We call this facility Machina East.

We're using this facility to manufacture our next generation, which is going to be portable.

Not only can we deploy it anywhere on Earth, but the portable nature of it can be also put eventually on a rocket and deployed to space.

AI is crucial to roboforming.

With artificial intelligence, now we can enter the realm where a robot can be flexible, can decide what to do and what not to do.

It can change the set of operations that it was planning to do, based on the inputs it's getting from the environment.

In the future, AI-guided robots could enable numerous kinds of manufacturing to take place in space.

A number of private companies have grand ambitions.

This company, Relativity Space, wants to 3D print entire rockets on the surface of Mars.

While another, Orbital Composites, is working on 3D printing solar power stations in orbit to test out space-based solar power.

It could take decades to launch some of these technologies.

However, the world's most well-known space agency is on board.

In addition to Artemis, an international effort for persistent presence on the moon, which is really focused on getting the crew on the lunar surface, there are supporting missions to supply the materials, the robots, everything that we need to be able to build that infrastructure, habitat, landing pads, roads, power stations.

I mean, think about building a small city on the moon.

But the moon will also serve as a pathfinder, a testbed, a place for us to learn as we transit to Mars and seek a permanent presence there as well.

Settling humans on Mars is a regular story in science fiction.

And many real people today, billionaires and beyond, have dreams to do it too.

The sophisticated AI and robotics in development today could help to build infrastructure that would make those dreams a reality.

We are treading new ground here.

We have not asked this much of our robotic systems ever before in the past.

I think one of the earliest applications will probably be in monitoring.

So imagine you've got habitats, even when humans aren't there, 24 hours a day, these need to run.

So you can imagine systems that are monitoring, doing fault detection, and maybe even prediction.

You could imagine even think about how humans work, handing off objects from one to the other.

All of this will be done robotically with the intelligence of the human brain.

The intelligence, AI, that we need for this agency.

AI-guided robotics could be used to construct space hotels for tourists, manufacture semiconductors or drugs in microgravity, and even mine asteroids.

As futuristic as all this sounds, AI and robotics are already being used in space.

Liftoff of the Delta rocket with Mars Pathfinder.

NASA's troop of Mars rovers has been trundling over the red planet since 1997.

AI was used to pilot the unmanned Deep Space One after it was launched in 1998 to investigate an asteroid and comet, while free-floating robots, called Astrobee, help astronauts aboard the International Space Station with their routine duties.

However, using AI and robotics to build in space presents plenty of challenges.

The robots will need to withstand the harsh conditions of space and operate in different gravity to Earth.

And it won't be easy to call out a mechanic if something goes wrong.

Mark Woods is a specialist in autonomous robotics and AI.

I think the biggest challenge we have in terms of doing things in space is just hard to get people up there if things go wrong.

The second thing is a lot of the communication we have with robots on planetary bodies is not in real time.

So, for example, it can take anything from 3 to 22 and a half minutes to get a signal from Earth to Mars.

It's just impossible to do what we might call teleoperation or joystick control of things.

That means that the robots that we have on Mars really have to have some element of agency and autonomy.

The technological and financial hurdles involved in creating such systems are huge, but so too are the possible rewards.

The vehicle has landed.

The global space business could generate revenue of more than $1 trillion by 2040.

Enhancing the capability to manufacture in space and to build cities there could bring economic benefits to humans.

Perhaps more significantly, it could help develop humans' understanding of their place in the universe.

Exploration is part of our nature.

I don't think we're going to sit idly on Earth and just send robots.

I hope that as humans we can also go there and experience those outer worlds to become a multi-planetary species.

Hello, I'm Alok Jha, science and technology editor at The Economist.

If you'd like to read more about AI's impact on science, then click on the link opposite.

And if you'd like to watch more of our Now and Next series, click on the other link.

Thanks for watching and please don't forget to subscribe.