AI saving the planet

Climate change is one of the most critical issues facing the world with a 46% increase in CO₂ levels since 1990, average wildlife populations have dropped by 60 per cent in just over 40 years and the world’s tropical forests are shrinking at the equivalent of 30 football pitches per minute. AI is one of the pioneering solutions to solve this climate crisis.

AI could be invaluable in mitigating and preventing the worse effects of climate change
— Google Brain co-founder Andrew Ng

One of the most effective solutions of climate change is to switch to renewable sources of energy such as wind power which do not release as much greenhouse emissions however they are not as reliable. Google is making this wind energy more viable using the AI software of its London-based subsidiary DeepMind. DeepMind’s machine works by learning algorithms to predict the wind output from the farms Google uses for its green energy initiatives. This will allow Google to schedule set deliveries of energy output rather than provide standard non-time-based deliveries to the grid. Allowing wind energy to be 20% more valuable and a more viable source of energy production.

We can’t eliminate the variability of the wind, but our results suggest that we can use machine learning to make wind power sufficiently more predictable and valuable
— Sims Witherspoon, a product manager at DeepMind
Showing the process of the DeepMind System - source: DeepMind

Showing the process of the DeepMind System - source: DeepMind

An alternate application of AI in helping the environment is by helping increasing sustainability by discovering materials faster than ever. For example, recently, researchers at Northwestern University used AI to figure out how to make new metal-glass hybrids 200 times faster than they would have by doing experiments in the lab. They do this by building databases of thousands of compounds so that algorithms can predict which ones combine to form substances with particular properties. Databases can also be used to discover exactly what makes a material so much e.g. stronger or lighter. Furthermore, AI has made calculations sophisticated enough to predict the properties of a possible new material on a computer before it’s ever made in a laboratory. This has allowed scientists to preserve precious raw materials increasing sustainability. But also discover revolutionary materials which can help combat climate change much faster.

Our goal is to discover more efficient and cost-effective ways of making materials that we already make
— MIT materials scientist Elsa Olivetti

These are just two of the thousands of many effective applications AI has in the future in combating climate change. However, AI is still in the very early stages in this sector and many challenges remain. Such as AI currently operate on more simplified models that don’t take into account real-world factors such as temperature and humidity.