in the way science is done.

Not only tackling the problem of climate change

and trying to find solutions,

but also actually trying to understand the system itself.

So some of the technological advances that we might be able to use

to combat climate change involve the ability

to generate energy in sustainable ways.

The question today is can we make solar panels or other materials

that are capable of generating energy in useful ways?

In the past, scientists and engineers have developed new materials,

but in rather a haphazard way.

Thomas Edison created the lightbulb

but he developed the original filament

by testing thousands of materials to find eventually the one that worked.

Today we can use machine learning and artificial intelligence

in our search for new materials.

The reason that we can do it better in the computer

is that the computer allows millions of potential solutions

to be searched in ways that you couldn't do in the lab.

The reason that AI is important in the development of our understanding

of the Earth's system is that we're getting more and more data

about the Earth as time goes on.

So today we collect satellite data,

remote sensing observations allow us now

to look at the patterns of landslides.

As climate is changing, as weather patterns change,

is the land being more destabilised by excess water?

Or are the erosion patterns in the mountains

changing as climate changes?

Extreme weather events as a result of climate change

do appear to be on the increase.

But there's another aspect of this - cities are getting bigger,

areas that have always been vulnerable to natural hazards.

With so many people potentially at risk,

it becomes important to understand the system very quickly.

Examples include understanding the ways in which

to respond to a situation.

Where to send the first responders.

Which hospital is put on high alert.

Where to send relief supplies, who needs tents.

So all of these sorts of disaster relief processes

depend upon information.

And we know, as we travel around congested cities,

that we spend a lot of time sitting in traffic,

burning energy, getting nowhere.

Autonomous vehicles provide routes

to making transport systems more effective, more efficient,

so that you don't have these waves of static traffic on a motorway.

That everything is moving at the right speed

and gets from one place to another in the most effective way,

the most efficient way.

That's not something that, as individuals,

we're very good at planning for ourselves.

So if we want to reduce, recycle, reuse, can AI help us?

Well maybe. But it depends upon our motivation.

AI is a tool, it's not a master, and so our responses ultimately

will depend upon our own personal motivations,

and those of the society that we're part of.

But AI and machine learning can help us move in the right direction.