What role does AI play in solving climate change?

Machine learning. Quantum computing. ChatGPT. We all have heard about these AI platforms and they clearly have so much potential. All the way from school essays to healthcare problems to encryption, AI will soon be the “new Google”. Right now, AI is in a relatively primitive phase. While it is impressive that you can ask any question to ChatGPT and it will spit out an answer, AI can be used for so much more than that. Far in the future, things like AI, machine learning, and quantum computing can be used for global issues, such as vaccine development, encryption, security, and even solving problems, such as New York traffic and space travel, as well as even larger problems like climate change. I’m sure you know where I’m going with this. What is AI’s role in solving the climate crisis?

There are a number of ways that AI can be utilized to help the climate crisis. To address them, first I am going to break climate change down. There are many causes of climate change, even more effects, and even more aspects that we need to consider when even thinking about solving it. Let’s start with the causes.

The main contributor to climate change is carbon emissions. We can break the emissions down into two main parts: inherent and human-caused. Yes, arguably, all emissions that are contributing to climate change are human-caused, but I’m going to redefine the words “inherent” and “human-caused”. When I say “inherent” I mean chemical or natural processes, such as creating cement: emissions that cannot be avoided because we don’t have the innovation to do it a different way. When I say “human-caused” I am talking about things like cars, factories, manufacturing, electricity; most of these things have a greener alternative — even if they aren’t as efficient or economical — the necessary technologies to replace them do exist. How can AI help?

Let’s start with the inherent emissions. Some examples of inherent emissions are cement production and synthetic fertilizer. In cement production, the inherent emissions come from burning limestone, which results in necessary calcium for cement and a byproduct, carbon dioxide. As of now, there is not another way to make cement that avoids this problematic chemical reaction. The second example that I mentioned was synthetic fertilizer. Synthetic fertilizer is fertilizer that was manufactured in a factory. Synthetic fertilizer is abundant in nitrogen, and it is atrocious for the environment for this reason. Nitrogen is necessary for plants, however, less than half of the nitrogen applied to farm fields is actually used by the plant. The remaining nitrogen pollutes water or, more importantly, air. It enters the atmosphere in the form of nitrous oxide; a greenhouse gas that has 265 times the global warming potential of CO2 and is significantly more detrimental to the environment. In a nutshell, synthetic fertilizer is terrible for the environment. There is no practical way to make cement or synthetic fertilizer greener for a few reasons. First, the heat, electricity, and gasoline used in the production and transportation cannot be replaced with their zero-carbon alternative because it would be so expensive to a point of impracticality. In terms of the actual chemical processes and the inherent emissions, as of now, we do not have alternative methods that are less harmful for the environment. We have one way to do these things, and that way inherently produces greenhouse gasses. We don’t have a greener way because that would take years of trial and error and large sums of money — money that people don’t want to spend because the current methods work so well. This is where AI comes in, specifically, quantum computing. Quantum computers work in a way that let them explore many, many possibilities at once. You can see where I’m going with this. In the future, once it is developed, quantum computing can explore the different possibilities and ultimately come out with methods to make cement and fertilizer which are economical and don’t harm the environment. It sounds wishful, but this is what quantum computing promises to do. Therefore, while it seems bleak, there is hope for finding ways to eradicate our inherent emissions.

So, let’s look at the human-caused problems. Like I said, human caused emissions consist of things that can be avoided, such as cars, factories, electricity, and consumerism. Overall, we have a lot of innovation in these areas. For instance, there are numerous forms of clean electricity, electric cars, and electric water heaters and furnaces. All of these things exist, they just need to be affordable, efficient, and scalable. Some of our current technologies check one or two of these boxes, but they need to check all three. In these situations, in which we have the innovation and green alternatives we need, there isn’t much that AI can do. Some of the solutions for problems with scaling new technologies would require illogical steps, such as extremely high taxes, reducing social service investments, like healthcare, and heavy, unfair regulations on the private sector. In these situations, it is up to us to adopt these technologies and place reasonable, yet stern laws in place that allow these technologies to become cheaper and more widely available. Many of the innovations that we already have can be scalable, if we do the right things in terms of the government, and private and public sectors, but those steps require human action, not AI insights. AI can tell us what to do, it can give us new technology, it can lay out a perfect step-by-step plan, but ultimately we have to be the ones to execute it.

Now let’s look at the effects. AI can help here. Climate change is going to cause dislocation, harsher, unpredictable weather conditions, increased drought, warmer temperatures and much more. Then, these initial effects cause a number of other things including loss of biodiversity, exacerbation of poverty and hunger, insect outbreaks, colder temperatures, and rising sea levels. AI can solve these issues in a number of ways. Recognizing weather patterns can help give early storm warnings and decrease human casualties. Organizing efficient methods of relocation and protection can help solve for a soon overwhelming number of climate refugees. Finding ways to protect crops from insects, floods, droughts, and heat is a way to prevent the exacerbation and cause of a new hunger crisis. These sound ambitious, but they are very real and probable things that can happen at a fast rate with the help of AI.

AI is very promising when it comes to the inherent emissions and effects of climate change, and in the future, it might even be able to solve some of the human-caused emissions. AI can help so many people who will suffer from the effects of climate change. Not now, but soon, AI, piece by piece, can help us solve the climate crisis. Of course, this promise is not one that means we can continue our daily lives and wait for AI to save the day. We must still continue and enlarge all our efforts to solve the climate crisis because while AI can give us solutions, it is ultimately up to us to carry them out. In the future, developed AI can pave the road for us. But it’s up to us to make it to the end.