Researchers aren't just advancing AI with no thought in mind, they hope to apply it in a variety of fields, and solve some of the most pressing problems of our time.
One of those where we can try to use artificial intelligence to our advantage though concerns the environment, and how to undo the damage we've caused.
A new research paper authored by a number of international scientists demonstrates just this -- even though there's evidence that AI development itself is contributing towards deadly climate change.
So here's how we can leverage AI to save the world.
1. Better electricity systems
Our electricity production and distribution facilities are flooded with data that we're just wasting right now. By using AI, we can forecast electricity generation and demand, in order to more efficiently integrate renewable energy resources into the grid and reduce waste
2. Monitor agriculture and deforestation
Greenhouse gases don't just come from cars and factories, a lot of it is also thanks to us destroying vital carbon-capturing woodlands, whether for construction, farming, or cattle. Using satellite imagery combined with AI, we can figure out where this is happening anywhere in the world, and control how much destruction we're causing.
3. Develop new low-carbon materials
According to the paper, nine percent of all global greenhouse gas emissions come from concrete and steel production plants. Machine learning can help here by giving us the tools we need to develop better, stronger materials, that are also carbon neutral at the very least.
4. Predict extreme weather
Thanks to climate change, a lot of the world is experiencing destructive weather conditions like hurricanes and heat waves. At the very least, we can use AI in tandem with current data gathering methods in order to better predict things like droughts, floods, and storms, allowing government to protect the people in their wake.
5. Refine transportation
At least 25 percent of all global CO2 emissions are thanks to the transport sector, two-thirds of it from road-users. Machine learning can help here by increasing vehicle fuel efficiency, and deploying shared transport systems (including autonomous vehicles) in order to lower our carbon footprint.
6. Reduce buildings' energy waste
Commercial structures waste a lot of energy, through central heating and cooling. AI here can identify the factors involved like sunlight direction, and how well floors are laid out, to optimise climate control.
7. Make the very Earth reflective
One of the more extreme applications of AI toward climate change involves a little bit of geo-engineering. The researchers say that machine learning can be used to find a way to make clouds more reflective, or even produce artificial clouds, in order to reflect back most of the heat coming off the Sun. Of course, we'd have to keep potential side effects in mind too.
8. Provide personalized carbon-lowering tools
The scientists say AI can be hugely beneficial when it comes to lowering your carbon footprint on a personal level. A simple algorithm can study your daily travel, food, and energy usage patterns in order to suggest more efficient routines. Add them all up, and it makes a big difference.
"Technology alone is not enough," they write, led by University of Pennsylvania's David Rolnick "Technologies that would reduce climate change have been available for years, but have largely not been adopted at scale by society. While we hope that ML will be useful in reducing the costs associated with climate action, humanity also must decide to act."