Artificial intelligence and the road to sustainable development
Artificial intelligence and the road to sustainable development
Throughout history and until today, humanity hasn’t gotten tired of trying to invent tools and devices that help it perform its daily tasks, which turned over the centuries into an attempt to find a way for these tools and devices to perform some tasks entirely instead of the human being, and then the matter developed into a pursuit of inventing a machine that could simulate the human mind in the way of thinking to find solutions to the most complex issues, hence the term “artificial intelligence”.
Human interest in the concept of “artificial intelligence” may have appeared timidly in several novels and stories in the nineteenth century and then in movies and theaters, all of which tried to draw imaginary images of how human and machine would merge, and some of them even predicted the dominance of those machines one day.
Although “artificial intelligence” or “AI” in its contemporary concept began in the fifties of the last century, this term did not gain the same momentum that it is enjoying now except in the last two decades, and this is due to the steadily increasing reliance on artificial intelligence technologies in many devices and various software applications, which led to a major technological revolution in many fields and a reformulation of our reality to be more modern, smart, and even “artificial”.
Like other modern technologies, artificial intelligence carries many solutions that support the achievement of sustainable development goals, from the accelerated development of the quality and efficiency of numerous services to enhancing various industrial processes by finding out innovative solutions to problems that humanity thought were eternal. On the other hand, artificial intelligence can negatively and sharply affect many human activities by replacing the human element with a machine, and this, of course, raises a lot of concerns.
In the pages of the featured article for this issue, we will try to provide a complete picture of artificial intelligence, including its advantages and disadvantages, to understand the potential of artificial intelligence and how it can represent the x-factor in helping us achieve the goals of sustainable development and ensuring a better future for us and the whole world.
Definition of artificial intelligence
The term “artificial intelligence” has witnessed, over successive decades, a change in its perception and how it is defined, as this term has become more comprehensive with time. At the beginning, we defined artificial intelligence as one of the branches of computer science concerned with the development and manufacturing of smart devices and machines that can conduct tasks that usually require human intervention.
Our deeper understanding of artificial intelligence, according to the variables of our current reality, has transformed our narrow perspective into a much broader one, due to the association of artificial intelligence with other terms such as “machine learning” and “deep learning”, among others.
So, we can say that AI is a scientific field that is concerned with building computers and machines that can think, learn, and act in a way that would normally require human intelligence or involve data whose scope is beyond what humans can process.
Artificial intelligence is a broad field that spans many different disciplines, including computer science, data analysis, statistics, hardware and software engineering, linguistics, neuroscience, and even philosophy and psychology.
In general, AI systems can perform tasks related to human cognitive functions, such as understanding words and their meanings or even playing games and identifying patterns. AI systems can do this by processing massive amounts of data and searching for different patterns that enable them to model their decisions.
For example, the three historic matches that brought together the world chess champion at the time, the Russian “Gary Kasparov” and the computer called “Deep Blue”, which was specially developed by “IBM” to play chess. The first match was in 1989, when the Russian player was able to defeat the computer; the rematch was in 1996 with a more powerful version of the computer, but that was not enough to defeat the greatest player in the history of chess.
Work at “IBM” did not stop after the last defeat, and only one year later, in 1997, the world waited for the third face-off between man and machine, but this time the result was different, as “Deep Blue” was able to defeat the world champion, which many considered a milestone in the history of artificial intelligence.
Despite its superior capabilities, human supervision is still required for AI systems because it enhances good decisions resulting from these systems and reduces bad ones. But some AI systems are designed to learn without supervision; for example, AI can play a video game over and over until it finally figures out the rules and how to win.
History of artificial intelligence
Artificial intelligence in its contemporary concept appeared in the first half of the twentieth century, when it made its way into science fiction films that introduced the concept of a robot that could think and make decisions.
By the fifties of the last century, we had a generation of scientists, especially mathematicians and philosophers, who believed that the concept of artificial intelligence is something that can be understood in both theory and in practice, and among those a scientist called “Alan Turing”.
Turing, who later became the “father of theoretical computer science and artificial intelligence, explored the mathematical possibility of applying artificial intelligence.
Turing believed that humans use available information in addition to the causes and circumstances surrounding each situation to solve problems and make decisions, and therefore, if we can transfer this method to machines, they can become smart enough to make decisions. This was the logical framework that Turing published in a 1950 research paper entitled “Computing Machinery and Intelligence”, in which he discussed how to build intelligent machines and how to evaluate their intelligence.
Unfortunately, Turing’s revolutionary theory did not find the needed resources to bring it into practice, as computers at that time needed a radical change. To make it clear, computers at that time lacked the basic requirements for intelligence. They could not store orders but just do them; in other words, computers can be programmed to do things, but they can’t remember what they performed previously.
The other thing is that computing was awfully expensive. In the early fifties of the last century, the cost of renting a computer reached two hundred thousand dollars per month, and therefore these devices were only available at prestigious universities and large technology companies, due to their ability to bear these costs, so convincing any investor to put their money into the development of this crazy idea was not easy at all.
Five years later, computer scientist “Allen Newell”, programmer “Cliff Shaw”, and engineer “Herbert Simon” designed the first program to mimic human problem-solving skills, a program that many consider the first born for artificial intelligence. This leap shook the scientific community and made it crystal clear that artificial intelligence is a theory that can be achieved, which opened the door for more research and studies.
The period between 1957 and 1974 witnessed a great advancement in artificial intelligence, thanks to the great development in the capabilities of computers to store more information, while becoming cheaper, faster, and more available. Machine learning algorithms also improved, and investments in this field increased dramatically and steadily.
This accelerated movement of research and development revealed many formidable obstacles in the way of artificial intelligence, the biggest of which was the computer’s lack of computational ability to do anything substantial. On the one hand, storage space was still insufficient, and on the other hand, the speed of data processing was still underwhelming. This led to a slowdown in research and development for about ten years.
Ironically, in the absence of government funding and public hype, AI thrived during the 1990s and 2000s, during which many of its notable goals were achieved. In 1997, IBM’s Deep Blue defeated world chess champion Garry Kasparov, and in the same year, speech recognition software, developed by Dragon Systems, was launched on Windows. This was another big step forward in spoken language interpretation by computers.
And now we are living in the era of “Big Data”, an era in which we have the ability to collect huge amounts of information that no one can process, and therefore reliance on artificial intelligence in this regard has already become fruitful in many industries such as technology, banking, marketing, and entertainment, especially with the revolutionary development in the capabilities of computers, whether in their storage or computational capabilities, which led to many breakthroughs in the field of computer science in general and artificial intelligence in particular.
So, what does the future hold? Soon, the language of artificial intelligence will be the primary language of the planet, and we see it clearly day after day in our daily activities. For example, when was the last time you contacted a company and spoke directly with a human being? Mostly, you talk with a smart assistant that helps you get what you want, and sometimes you find that someone calling you is a voice program that conveys a promotional message for a product.
With the development of artificial intelligence capabilities, it will not be strange that, in the future, we will have full conversations with artificial intelligence programs.
Also in the future, we can expect to see self-driving cars. This technology is already being introduced in the form of advanced features in our vehicles, such as automated parking and lane change assist, all without your hand touching the steering wheel.
The biggest goal for the development of artificial intelligence is “general intelligence”, by which we mean that machines and programs have cognitive capabilities that surpass humans in all tasks.
Types of artificial intelligence
Artificial intelligence has created new opportunities for improvement and development in key sectors such as health, education, and the environment, thanks to its ability to perform some tasks more efficiently or systematically than humans.
For example, we find that smart technologies in buildings and various vehicles contribute significantly to critical issues such as reducing carbon emissions and supporting people with disabilities. In another context, machine learning, a branch of artificial intelligence, enabled engineers to build robots and self-driving cars, among other applications.
We can divide artificial intelligence into three main categories according to its ability to learn: weak or narrow artificial intelligence, general or strong artificial intelligence, and super AI.
These categories show the capabilities of artificial intelligence through the curve of its development during the execution of precisely defined sets of tasks, which require the same ability to think as humans, or perform tasks that exceed human capabilities, but we will not elaborate on these three categories; instead, we highlight the classification of the researcher and professor of integrative biology at Michigan State University, “Arend Hintz”, who divided artificial intelligence into four main types.
First: Reactive Machines
Reactive machines are artificial intelligence systems that do not contain memory and are task-specific, which means that data input to these machines means output of the same nature and type. We find this type in machine learning models, as they take customer data such as purchase history or search history and use it to provide appropriate recommendations to the same customers, which we all experience when browsing the Internet.
An obvious example of this type is the IBM chess computer that we mentioned earlier. This computer can recognize its own and its opponents’ pieces on the chessboard to make future predictions, but it does not have the memory to use past mistakes to make future decisions; it only predicts the next possible moves and selects the best one; thus, reactive machines can be considered a model of super artificial intelligence.
Second: Limited Memory
The next type of AI is limited memory, a type that mimics the way neurons in our brains work, which means that systems of this type get smarter over time as they receive more data to train with.
Unlike reactive machines, limited memory can investigate the past to remember certain things or situations over time. These observations are then programmed into AI systems so that they can perform their actions based on past and present moment data. However, in limited memory, data is not saved for the long term as an experience to learn from. An example of this is some self-driving technologies that allow a vehicle to change routes according to information gathered during previous trips.
Third: Theory of Mind
The first and second types of artificial intelligence are the two types that currently exist on the ground. As for the third type, “theory of mind”, and the fourth type, “self-awareness”, they are among the types of artificial intelligence that will be built in the future, and therefore there are no examples of them.
If it is developed, theory of mind AI will be able to understand the world around it more deeply, absorb thoughts, emotions, and beliefs, and thus determine how it behaves or deals with them.
We humans understand how our thoughts and emotions affect others, and how other people affect us; this is the basis of human relationships in our society, but in the future, AI machines of this type may be able to understand intentions and predict behavior as if they were simulating human relationships.
Fourth: Self-Awareness
Self-awareness as a concept goes beyond the theory of mind and understanding emotions, whereby a system or machine is aware of itself and could sense or predict the feelings of others. For example, “I’m thirsty” becomes “I know I’m thirsty” or “I want to eat eggs because it’s my favorite food”.
The ultimate stage of the development of artificial intelligence will be the design of systems and machines that have a sense of themselves and an understanding of the reason for their existence, and this type also does not exist on the ground yet.
We are so far away from self-aware AI because, even now, we humans are still learning more and more every day about how our brains work in learning and decision-making.
The role of AI in development
Artificial intelligence has the potential to solve some of the biggest challenges facing humanity, something we see today in every sector. Around the world, we are already witnessing AI-derived growth in many fields. For example, in the car industry, advanced driver assistance systems (ADAS) are now introduced in many vehicles with a set of features that significantly improve vehicle safety and accident avoidance.
In the healthcare sector, using advanced AI technologies, biomedical researchers are increasing the pace of drug discovery by identifying chemicals, molecules, and compounds that may be better for specific diseases. On the other hand, AI technologies allow financial services companies to better identify fraudulent or corrupt activities.
We cannot forget here the important role of artificial intelligence in combating climate change, by improving the efficiency of various industrial processes and reducing waste in resources. In addition, AI can predict weather patterns and wind movement to combat pollution more effectively.
Consequently, we can say that artificial intelligence is the ideal ally that humanity needs to plan and implement different strategies to achieve sustainable development goals more effectively.
Technologies like artificial intelligence will help us build more efficiently, use resources sustainably, and reduce and manage the waste we produce more effectively, among many other things.
The combination of artificial intelligence and sustainable development will help develop various industries in a way that allows our human civilization to advance better, while meeting current needs without compromising future generations, especially considering the critical issues that we are facing today, mainly climate change and its consequences.
According to a study published in Nature, AI can help achieve most of the Sustainable Development Goals (SDGs), as this technology can become a key tool to facilitate the adoption of the circular economy and the building of smart cities that use their resources efficiently.
A clear example of the contribution of artificial intelligence to sustainability is traffic management, where artificial intelligence allows urban mobility to predict traffic jams and suggest alternative routes, thus reducing fuel consumption. AI can also predict times of increased demand for transportation, enabling transportation companies to schedule their trips appropriately while making it easier for passengers on the one hand and reducing the impact on the environment on the other.
Artificial intelligence can also help enhance the efficiency of renewable energy plants, which is a practice already adopted by many companies nowadays. Through artificial intelligence, it is possible to prevent and diagnose power plant failures, in addition to predicting the quantities of energy demand in the coming days and adjusting those plants to meet these demands more efficiently.
Outside of the energy sector, there are many industries and companies that could improve thanks to AI. In agriculture, for example, AI is being used to make irrigation and fertilization more efficient thanks to moisture, temperature, and fertilization sensors. AI can also predict crop needs.
One of the most innovative solutions that AI provides for sustainable agriculture is drones. These drones can help farmers with continuous monitoring of large agricultural areas, in addition to analyzing images and using the analysis results for comprehensive pest control.
For industry in general, AI provides more efficient warehousing, manufacturing, and distribution systems. Moreover, artificial vision systems also make it possible to identify errors in assembly lines that are invisible to the human eye, as well as potential safety risks or disasters. Safety is one aspect that is particularly important in industries such as construction, where safety is a paramount necessity.
Another example of using artificial intelligence to achieve greater sustainability is the technology developed for tunnel boring machines. These machines are particularly complex. During the drilling process, any landslide can stop all or a large part of the underground work, but here comes the role of AI.
Oracle was able to develop a prediction system that processes continuous information for 3000 variables to predict the occurrence of any error during the drilling process, and thus the economic savings and efficiency in tunneling will be greater than ever before. Therefore, artificial intelligence is an essential element in terms of predicting errors and reducing problems that may affect sustainable development.
AI and the environment
AI is now playing a role in manufacturing, education, and many other fields, but can AI help us save the planet? How can artificial intelligence contribute to addressing climate change, and how can it negatively affect the environment? All these questions are discussed in the following points:
1- Climate monitoring
We already have a lot of equipment for monitoring our climate, but incorporating AI into such systems could allow us to improve our monitoring capabilities. With the help of artificial intelligence, certain weather patterns that were not observable using traditional methods can be identified. Moreover, AI can draw important inferences about the direction our climate is headed, allowing us to better mitigate negative impacts.
To achieve this, it may be useful to equip orbital satellites with AI capabilities so that a larger picture of our planet’s status can be drawn. AI climate monitoring can also be used on a smaller scale to assess the changing nature of a single ecosystem.
2- Monitoring the carbon footprint
Carbon emissions are a major contributor to climate change. “Our World in Data” reports that carbon dioxide has been emitted at a rate of 1.5 trillion tons since 1750. It is clear to most of us that tackling climate change largely involves reducing carbon emissions, but this has not happened or is happening at a slow pace.
Using artificial intelligence, the carbon footprints of large corporations, such as oil and gas companies, clothing manufacturers, and giant agricultural entities, can be continuously tracked to determine the impact of their operations on our planet. In addition, AI can calculate the carbon footprint of individuals or an entire population, such as the residents of a major city or small town.
The use of predictive AI (i.e., data analysis and machine learning to make predictions) may also play a role here, as it can predict future carbon emissions, allowing us to see how things are going based on our current activities, which in turn can help create environmental laws and regulations that are imposed by governments as well as international and regional agreements.
Another interesting potential use of AI in mitigating the global carbon footprint is to identify where the large sources of carbon emissions come from, highlighting the main contributors to the problem. This can be done broadly or narrowly, but either way, it can provide us with some basic information about what or who is harming our planet.
3- Monitoring changes in land uses
Today, our planet’s land use plays a major role in climate change, with informal agriculture, deforestation, and urban sprawl contributing to global warming through carbon emissions and the removal of natural habitats. For this reason, it is important that how land is used to be constantly and carefully monitored.
In this approach, we can utilize AI to understand how we may improve our land-use practices and possibly get new ideas on how to manage land more effectively. For example, parts of desert may be used to build solar panel farms, or coastal areas could be used to harness tidal energy.
4- Generating innovative green ideas
Because artificial intelligence is a type of intelligent technology that can learn and correlate information, it can be effective in producing new and novel solutions to combat climate change. For example, artificial intelligence can be provided with data about how to use various sources of energy, as well as the technology accessible in our modern time and the environmental goals that we want to achieve. After that, the AI system can list some options for producing energy in a more intelligent way that can reduce the use of resources, reduce carbon emissions, and even save money.
AI can also be used at the end-user level to save energy in homes, or give businesses more efficient ways to operate machinery, move goods, and more. AI can also be integrated into applications that focus on the environment so that users can find new ways to reduce their carbon footprint in their daily lives.
While we already have human experts who can help in this area, the sheer amount of data that AI systems can process and their ability to notice patterns and search for solutions make them a promising candidate in our quest to be kinder to our planet.
5- Prediction of environmental disasters
One of the frightening consequences of climate change is the increased frequency of natural disasters, such as floods and forest fires. These events can be devastating and can take many lives, so it is crucial that we have technology to monitor the planet and alert interested parties of any imminent threats.
AI can be used in sensory equipment, monitoring equipment, and similar technologies to learn the natural patterns of different environments and then sound an alarm if something changes. For example, AI can monitor weather systems, predict precipitation, warn of high temperatures in forests, and much more.
In addition to the above, AI may have the ability to devise effective plans for evacuation, rescue, or rebuilding in the event of a disaster. For example, AI can be used to find suitable evacuation routes, and it can help architects build more disaster-resistant infrastructure to mitigate damage should it occur.
6- The negative environmental aspect of AI
Although there are many ways in which AI can help us combat climate change, this technology, like others, is already harming the environment.
Developing and testing AI systems take significant amounts of time and energy, with reports indicating that a single AI model can emit up to 300kg CO2eq, which is the same as the total CO2 emissions of five cars.
It is important to note here that despite all the potential benefits, the development and creation of AI systems has a large carbon footprint, which is a downside that we should neglect when discussing the great benefits and huge potential of AI in our future.
But what about the ethical side of artificial intelligence? Can it be transformed from a tool of development into a new tool of control and exploitation in the hands of developed countries? We leave the answer to these questions to Dr. Muhammad Al-Khayyat, Chairman of the “New and Renewable Energy Authority”.
The mighty AI – Dr. Muhammad Al-Khayyat
We were discussing a project situation when I highlighted the importance of emailing the construction company; to reflect the position of the institution, I mentioned several points to be included in the message. In less than a minute, a female colleague pointed out to me that the mail proposal was ready to be shown on the screen in front of the audience for discussion. The text was largely acceptable except for some minor remarks, and when I asked her, “How, so quickly?” She replied, “It is the mighty AI”.
During the rule of the “German People’s Police” in East Germany, before the collapse of the Berlin Wall in November 1989, and for the purpose of obtaining high-level security reports, tens of thousands of citizens were appointed secret informants, which led to the filing of thousands of reports that needed to be carefully analyzed to draw the right conclusions. At that time, the officials struggled to even read most of these reports and started facing many questionable decisions.
Today, the digital revolution is taking care of our information, as phones and computers send streams of data about the details of our lives to data centers around the globe. As a result, our lives have become a laboratory for human behavior and a field of experiments for the social, economic, and psychological sciences. Obtaining data has become possible and easy, and manipulating it has also become easier than ever.
For instance, we find cameras in our streets, offices, and bus stops, as well as smartphone cameras, which are nothing but eyes that sneak at every movement and moment of our lives, even the microphone in our phones is nothing but an ear that hears all our conversations and interactions. So, artificial intelligence became like the modern informant, who sorts, analyzes, and makes decisions. You can say that the big brother in the novel “1984” by the English writer “George Orwell” has become real!
As soon as the cashier processes a customer’s credit card payment, all the customer’s demands and desires are revealed, and thus companies determine the products that are most popular, the size of their stock available in the markets, and when to request additional shipments, as well as broadcasting marketing messages that differ according to the target group.
To expand the base of brand loyalty, some stores use smart shopping carts, which guide customers through a permanent customer card -a card that stores give to their regular customers- to discounted products with some quick tips that suit their desires.
All these services and others would not have been possible without artificial intelligence algorithms that rely on collecting and processing as much data as possible, which has become known as “big data”.
In 2018, Mark Zuckerberg, the former CEO of Meta -Facebook- was grilled by the US Congress over data misuse (data breach scandal) of millions of users by Cambridge Analytica.
The scandal has led to the resignation of its CEO, “Alexander Nix” at the time, especially after discovering that the company used some of this data to blackmail politicians and celebrities; in order for them to pay sums of money in exchange for the company’s silence on some of their actions, this is in contrast to the use of data to know voter orientations and political affiliations, with a focus on who have not yet decided for whom to vote; thus, selecting messages and advertisements directed at them in order to direct their affiliations.
Not only that, but predicting the trends of stock markets, stocks, currency exchange rates, and futures contracts -such as oil prices- has become dependent mainly on predictions based on scientific foundations.
The same is applied in medicine for the purpose of knowing the patient’s health condition and the diseases he went through, as doctors can predict the occurrence of some diseases. Thus, early detection reduces their effects, if not helps to avoid them.
Consequently, we can say that uses of artificial intelligence have grown in all aspects of life, but with a mixed bag of influences. As long as there is historical data and simulation programs, predicting the future is just one click away!
All the above are real and diverse models that answer the question: How can our future actions be predicted and how it can become the product of simulation programs that process our memories and past? The answer for this question is fueled by our curiosity to explore the technology to find out that how capable it is.
We must remember that when we express our concern about artificial intelligence, we are referring indirectly to us humans who program those electronic circuits to gain control (using special software) over some mechanical parts, which we will later call “robots”. Even though our intentions look innocent, I can’t hide my concern that this robot may one day become a wolf in sheep’s clothing!
