AI Ethics in Engineering
The Bias of Traditional Engineers in AI-based Modeling of Physics — PART 1
Outline
Ø Summary
Ø Introduction
Ø Data: Foundation of AI-based Modeling
Ø AI-Ethics addresses the Bias in AI-based Modeling
Summary
It is a well-known fact that predictive models developed by Artificial Intelligence and Machine Learning algorithms are based on “Data”. Since it is known how data is used to build AI-based models, the main characteristics of the AI-Ethics are about addressing how AI models become biased, based on the quality and the quantity of the data that is used during the model development.
When it comes to the non-engineering application of AI and Machine Learning, it has been proven that human biases such as racism and sexism can be included in AI models through the inclusion of biased data during the training of the machine learning algorithms. Since engineering application of AI and Machine Learning is used to model physical phenomena, AI-Ethics can determine and clarify how human biases of traditional engineers that include assumptions, interpretations, simplifications, and preconceived notions can be used in the engineering application of Artificial Intelligence and Machine Learning.
Introduction
The main reason that nuclear weapons did not end up destroying our planet (at least till now) had to do with the worldwide treaties and agreements on how to handle nuclear bombs. It is important that a similar set of worldwide treaties and agreements be eventually achieved by the politicians around the world about Artificial Intelligence. One of the main reasons that make many individuals to be worried about how Artificial Intelligence is going to impact our world in the next few decades has to do with the governments of several countries. Governments in some countries are using this technology based on their own objectives that are a function of their views, believes, and understanding of democracy and their intention of becoming the world leader based on how Artificial Intelligence can serve them. The Ethics of Artificial Intelligence has lately become an important topic that must be well understood by individuals that already are, or currently becoming interested in Artificial Intelligence and Machine Learning algorithms.
Since the mid-2000s that AI-based image recognition, voice recognition, facial recognition, object recognition, and autonomous vehicles were exposed to most people around the world, interest in Artificial Intelligence and Machine Learning has significantly increased. As new science and technology, AI and Machine Learning will change a lot of things in the 21st century. AI has become one of the most interesting technologies that people, companies, and academia are getting involved with, on a regular basis.
For example, recently, banks have started using AI and Machine Learning models to make the first step of the decision making about giving loans to applicants, while Human Resources of large companies use AI and Machine Learning models to make the decision about who to hire. From engineering point of view, some operating petroleum companies have been interested in using AI to develop fact-based reservoir simulation models.
Banks use the AI models to minimize the number of applicants that they must evaluate their characteristics in much detail while companies use AI-base models to evaluate the large number of applicants that have applied for employment based on the company’s job advertisement and then significantly reduce the number of applicants that the actual Human Resources professionals must concentrate on. Petroleum companies’ objective of using AI-based reservoir simulation is to enhance their oil and gas production. The way Artificial Intelligence and Machine Learning have been used by banks and companies to loan or to hire individuals has made the Ethics of Artificial Intelligence an incredibly important topic to be understood. The same is true for petroleum companies about the engineering application of Artificial Intelligence and Machine Learning.
The Ethics of Artificial Intelligence is important to engineers and scientists that have become enthusiasts to use this technology for solving engineering-related problems. While in the past several years AI-Ethics has become an important topic in the non-engineering application of Artificial Intelligence and Machine Learning, this article will explain the importance of Ethics of Artificial Intelligence in the engineering application of this technology. Specific examples of AI-Ethics in the engineering application of AI and Machine Learning are presented in this article. While AI-Ethics in engineering may not have much to do with politics (at least in this article) it is impacted to a large extend by (a) lack of scientific understanding of Artificial Intelligence, (b) lack of success in realistic problem solving through engineering application of AI, or © incorporation of traditional engineering biases (including assumptions, interpretations, simplifications, and preconceived notions) into the AI-based models of the physical phenomena.
Currently, some individuals and companies that claim they use engineering applications of this technology are including a large amount of human biases so that they can solve problems using Machine Learning algorithms after they fail to build an AI-based model that does not include human biases. Human biases in engineering have much to do with how mathematical equations are built to solve physics-based problems.
Data: Foundation of AI-based Modeling
Artificial Intelligence uses Machine Learning algorithms for developing tools and models to accomplish its objectives. The development of AI-based models has a lot to do with “Data”. The quality and quantity of the Data is the major impact of how the AI-based model will behave. As it was mentioned in the last section, banks have started using AI and Machine Learning models to make the first step of the decision-making about giving loans to applicants. The AI models are usually developed using historical data provided by the loan applicants along with previous results of loan payments. The amount of positive and negative loan payments as well as the input data from the loan applicants such as gender, ethnicity, credit, living location, income, etc. will determine the quality of the AI-based model that is developed for the bank loan. Such models can also include certain characteristics determined by the Bank management.
The same general approach is also applicable to the AI models for the Human Resources of large companies to make the decision about who to hire. Such models are also developed using existing data from multiple companies about the applicants as well as the quality of the employees that have been hired in the past. Other applications of AI that make AI-Ethics highly important include Face Recognition, Face Detection, Face Clustering, Face Capture, Face Match, etc. Such technologies are used by mobile phones, security, police, airports, etc.
In the Engineering application of AI, the characteristics of the data including its quality and quantity that is used for model development impact the quality of the AI-based models. Engineering application of Artificial Intelligence and Machine Learning is the use of actual measurements and actual physics-based data to model physics rather than using mathematical equations to build models for the physical phenomena. Traditionally, in the past few centuries, modeling physics at any given time had to do with engineers’ and scientists’ understanding of the physical phenomena that were being modeled. As scientists’ understanding of the physical phenomena enhances so do the characteristics of the mathematical equations that were used to model that physical phenomenon.
AI-Ethics addresses the Bias in AI-based Modeling
The characteristics of the quality and quantity of the data that is used to build the AI-based model determine whether any biases have been incorporated in the AI-based model. AI-Ethics’ objective is to identify the quality and quantity of the data that is used to build the AI-based model and to identify if any bias has been (intentionally or unintentionally) incorporated in the model through the data that has been used to build the model.
The way Artificial Intelligence and Machine Learning have been used by banks and companies to loan or to hire individuals has made the Ethics of Artificial Intelligence an incredibly important topic to be understood. The same is true about the engineering applications of Artificial Intelligence and Machine Learning. As long as realistic and non-traditional statistics-based machine learning algorithms are incorporated, the quality of the AI-based models is purely based on the quality and the quantity of the data that has been used to build the model. Therefore, the data that is used to develop the AI-based model completely controls the essence of the model that is developed and used for decision making.
As this technology moved forward and start solving more problems, scientists became interested in learning more details about how AI and Machine Learning work. It becomes quite clear that the main characteristic of AI and Machine Learning is its use of data to come up with the required solutions and to make decisions. Since data is the main source of AI-based model development, it became important to learn (a) where the data is coming from and what is the main source of it, and (b) to what extent the data includes all the required information (even not explicitly) that AI and Machine Learning can extract patterns, trends, and information from.
It took almost a decade of research and study until it became quite clear through examining the actual application of this technology that AI and Machine Learning have the potential of being political (Crawford 2019, Lim 2020), racist (Doshi 2018, Dave 2021), and sexist (Dastin 2018, Dave 2021). This has to do with the type of data that is used to build the AI and Machine Learning models. In other words, it is quite possible to create a biased AI and Machine Learning model that can do what you want it to do. It completely has to do with the data that is used to train and build the model. This is how AI-Ethics addresses the engineering application of AI when traditional engineers intentionally, or unintentionally, modify the quality of the AI-based models so that it would generate what they believe is the right thing rather than modeling the physical phenomena based on reality, fact, and actual measurements.
MIT’s AI-Ethics has published articles regarding the biases that can take place when using AI and Machine Learning. In some of these articles, it is clearly mentioned that “Three new studies propose ways to make algorithms better at identifying people in different demographic groups. But without regulation, that won’t curb the technology’s potential for abuse,” (Hao 2019–1), and “this is how AI bias really happens — and why it’s so hard to fix. Bias can creep in at many stages of the deep-learning process, and the standard practices in computer science aren’t designed to detect it.” (Hao 2019–2).
In another interesting article, it is mentioned that “Collecting the data; There are two main ways that bias shows up in training data: either the data you collect is unrepresentative of reality, or it reflects existing prejudices. The first case might occur, for example, if a deep-learning algorithm is fed more photos of light-skinned faces than dark-skinned faces. The resulting face recognition system would inevitably be worse at recognizing darker-skinned faces. The second case is precisely what happened when Amazon discovered that its internal recruiting tool was dismissing female candidates. Because it was trained on historical hiring decisions, which favored men over women, it learned to do the same,” (Terence 2020). What has been mentioned in this article is the results of research that was done to learn how bias can be included in the model. This is so true and very important for both engineering and non-engineering application of AI and Machine Learning. In this article, it will be shown how similar activities take place in the engineering application of AI and it will be explained in the next section what is bias when AI is used to model physical phenomena.
By doing some serious research on the fundamentals of AI and Machine Learning algorithms it becomes quite clear that this technology has an incredibly strong power of discovering patterns in the data that is used to train and develop models, make predictions, and help decision making. Since what AI and Machine Learning algorithms do, is all about data, then it also becomes clear that as long as the data that is provided to the AI and Machine Learning algorithms are generated based on biases, interpretations, and assumptions then the models and workflows that this technology develops becomes a representative of such biases, interpretations, and assumptions.
Pare 2: -> https://shahab-mohaghegh.medium.com/ai-ethics-in-engineering-437ec07046a6?postPublishedType=initial
