Author: Archit Diwakar, I Year of B.A.,LL.B from Rajiv Gandhi National University of Law, Patiala.
Abstract
Artificial Intelligence has become a necessary evil in the developing world. It is the main component of the Fourth Industrial Revolution. The blog deals with some of the major negative and positive impacts of the AI. The amount of energy that the algorithm to develop AI take is huge and the energy required to keep the data center cool is also tremendous.
The production of electronic devices and their transportation is also a major constituent of climate change. There are some positive impacts too, AI can help the farmers in the agriculture work, it can help the city planner and government to maintain the city and traffic properly and also can save the lives of people with the early warning of natural disasters. This blog is an attempt to have some insight about all these issues and benefits, and to bring some responsibility about climate change.
AI and Its Impact on the Environment
AI or Artificial Intelligence can be termed as the most impactful and pervasive technology of the Fourth Industrial Revolution. We can understand it in this way, of all the technologies in the Fourth Industrial Revolution, AI is its driving force or we can say it is the “electricity” of this revolution. AI has started to change how we live, work, and solve different challenges. From finding a grocery shop near us to helping in protecting wildlife and the environment, AI has very broad applications. But all these benefits come with a cost that the environment has to pay. So, we can say that AI has both positive and negative impacts on the environment.
Negative Impact of AI
AI technology now is being developed to address sustainable development goals. However, if we research a bit about this or start reading on this topic, we will find that there is great application yet often a lack of awareness of carbon footprints. At present responsible use of technology is very important for our shared ecology. In the year 2025, one-fifth of the global electricity could be consumed by our devices and the data we store; this is a pretty big slice of our energy.
Each app on our phone or service we use has data and this data has to be stored in a physical place on servers, you most likely not know where your data is, still our digital footprint is there, it is saved, trashed, archived. For example, “take emails, if every person in France will delete 50 of their emails right now, we can save enough energy to power the Eiffel Tower for 42 years and this is just the tip of the Eiffel Tower.”
Self-Driving Cars and Electronic Products
Being Ethical for a technology company means asking you who the self-driving car should hit, however, to me this is hit and run ethics. Certainly, if we go further, training one decision-making process on a lot of data could consume the lifetime emission of 5 normal cars. We can argue that self-driving cars would enable a better flow in traffic yet we have to remember the context. What will happen with a million self-driving vehicles? Currently, a large proportion of energy used in the technology is not renewable, we cannot easily recycle or reuse our electrical products, but if we try to establish a chronology, what industries pollute the most? The most common answer will be Mining, oil, or shipping, and all these are required industries to build and distribute the technology. Electronics after all are built through minerals and that means mining across the world. Minerals are transported around the world and fossil fuels are used in most cases to run this transport. When electronics are built, they have to be transported, manufactured again transported, and manufactured to a variety of locations across the world.
Cloud and Artificial Intelligence
Once these electronic devices are operative, they must be run through the cloud. Some technology leaders got their heads in the cloud, each cloud solution requires data and that means a physical data center. The material cost of building a cooling system for servers and elsewhere has one of the highest adverse impacts on the planet. When we say don’t worry just put it in the cloud that means a physical place that must be attended to. At the same time, the price of selling data has risen to the price of oil. It was said ‘data will become the new oil’ well it has, this is not something the technology companies should be proud of.
Now if we talk of AI, the adverse impact of running some AI models for years, it’s hard to come by, however, we know streaming videos accounts for one of the biggest chunks of the world’s internet traffic.
Technology companies are not open or direct about this matter and some actively work against the understanding of their impact on the environment. Let us consider a lesson from the past, the tobacco industry was funding scientific research on why tobacco is good for you in the same way the technology companies are funding scientific research on why AI is good for you, these technology companies want to find ways to measure and report their emissions. Technology companies are reporting on how climate-friendly their technology is not a good idea, this is the same as mining companies reporting on river spills and oil companies reporting on oil spills, it becomes unaccountable and technology leaders are not held responsible.
Large technology companies like Google, Amazon, Facebook, and Apple, do set targets to run on a hundred percent renewable energy, however, they have lost the trust of many with their use of private information and have been irresponsible in many cases. So, these companies should not be trusted to police their emissions.
I am not against technology but if we don’t solve the climate crisis with unlimited data plans, then what is the use of it, I am not against the use of Artificial Intelligence but the usage has to be in a more responsible way. We need systematic change and we have to change because we are the cleanup generations, it includes all of us. We have to build technology for the regeneration of our ecological environment. We can aspire to build Planet Center Technology. Let us make technology responsible. We can be making AI work for our shared ecology. We have to hear the voices of young people in the streets carrying signs, racing the planets, asking to be heard “you know it’s time for change when children act like leaders and leaders act like children.”
Positive Impact of AI
AI is evolving exponentially and so do the environmental problems. Scientists are now turning to AI to find sustainable solutions. If we see around us there are many ways in which AI can help us.
Agriculture
With increasing global warming, the temperature of the earth is also increasing and this increase in temperature has a significant impact on agriculture as well. But thanks to AI which can help in increasing the crop yield in these harsh conditions also. We can put sensors in the field to collect data about crop moisture, soil composition, and temperature, and transferring this data to AI will let us know when crops need watering. “Incorporating this data with drones, which are used to monitor fields, can help AI systems to know the best time to sow seeds, spray medicines, and harvest crops. This will lead to increased efficiency, increased yields and less use of water, fertilizers, and pesticides.”
Weather Prediction
What do we consider the most precious thing in this world? The answer most of you probably would give is ‘our life’, we can say that AI helps in saving this precious thing. If we start getting earlier warning signs of these catastrophic events then damages to human lives and property can be reduced by a significant number.
“There has been significant progress in using machine-learning algorithms that were trained on data from other extreme weather events to identify tropical cyclones and atmospheric rivers.” If the AI provides an earlier warning to the governments and the people, they will be in a better position to protect themselves. The weather forecast agencies are also deploying machines that can tell us how long the storm will last and its severity. Climate change is a huge problem and natural calamities are also increasing because of that. Therefore, we should focus on making AI and machine learning faster and smarter so that we can reduce the damages caused by climate change.
Cities and Traffic
With the growing population and urbanization, the cities energy consumption is increasing continuously. This condition can be tackled with the use of AI, with the help of the Internet of Things, sensors, and devices embedded in everyday objects to send and receive data, we can incorporate data from smart meters and allocate the energy and resources accordingly. With these meters, we’ll also know if the resources are being wasted that can be stopped. In big metropolitan cities, the problem of traffic is something that led to a loss of huge resources and also impacts the environment, but with the advent of AI, we can see the coming of AI-enabled self-driven electric cars. These cars would be inter-connected and would know the traffic conditions beforehand, which would optimize the traffic flow.
AI has many other uses as well; it can help in monitoring our wildlife and their ecosystem with the help of this we can prevent poaching and illegal logging in forests. AI can also be used to monitor the quality of residential drinking water, detect any leakage in the underground supply system, and can also suggest when does the water plant needs maintenance.
References
AI and climate change: The mixed impact of machine learning (techtarget.com)
World Economic Forum, ‘Harnessing Artificial Intelligence for the Earth (2018).’ Available at, Harnessing_Artificial_Intelligence_for_the_Earth_report_2018.pdf (weforum.org).