Introduction
People often think of robots when they hear anything to do with artificial intelligence. But is AI the same as robots? The Oxford Dictionary defines artificial intelligence (AI) as “the theory and development of computer systems able to perform tasks normally requiring human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages.” The origins of AI can be traced back to the 1950s, when researchers first began to explore the possibility of creating machines that were closely similar to humans (Su, 2018). The field of AI has gone through several phases of development over the past few decades, including artificial narrow intelligence (ANI), artificial general intelligence (AGI), and, most recently, artificial super intelligence (ASI). With the rise of big data, cloud computing, and advancements in processing power, AI programs have taken a big part in various fields, including healthcare, transportation, education, finance, and production, and have been successfully able to outpace human abilities in many of the technical skills.
Although AI successfully boosts productivity and efficiency across many fields, it also gradually led to various ethical concerns, including unemployment, inequality, and security. Since the applications of Artificial intelligence are seen in the everyday scenarios of our lives, it is crucial to know, understand, and explore how AI can pose threats to us in different aspects. This research paper will provide an in-depth analysis of the rise of ethical concerns due to the spread of Artificial Intelligence applications across various fields. The paper will argue that Artificial Intelligence has led to an arise in many ethical concerns which could potentially harm humanity more than it can ever benefit it.
Unemployment
New technologies are increasingly being adopted in all domains, including manufacturing industries, schools, and hospitals. People argue that AI will create more jobs, while others claim that it will completely replace workers. However, history shows that the broad adoption of technology causes widespread unemployment. According to Korinek & Stiglitz (2018), AI is a continuation of the automation process that started in the late nineteenth century. In each automation episode, technology performed the work that was initially executed by humans. For instance, there were numerous job losses between the 60s and 80s due to outsourcing and automation, and Youngstown, Ohio, lost around fifty thousand jobs within five years (Su, 2018). As a result, proponents posit that much of human labor is at risk of becoming obsolete and replaced by AI in all fields in the future.
Nevertheless, previous research shows that the risk of unemployment varies depending on the intensity with which AI is used in a workplace compared to the number of workers. For instance, about 47 percent of jobs in the U.S. labor market are at high risk of being replaced by technology over the next ten to twenty years. On the other hand, this risk is higher in the Republic of Korea, where approximately 55 to 57 percent of jobs are at risk of replacement (Lee et al., 2022). This insinuates that job insecurity will be higher in some places than others. Moreover, some jobs are at a higher risk than others, and this will widen the gap between skilled and unskilled laborers.
Bordot (2022) conducted a study to investigate the relationship between AI and unemployment, using data from 33 OECD countries. The results of these studies confirmed that AI tends to increase unemployment. The study also found that people with a medium level of education are the most affected by AI with regard to unemployment rates. Researchers claim that workers will be forced to adjust their skills in order to maintain their jobs because “the more comprehensive employees’ knowledge of AI, the lower their job holding insecurity” (Liu & Zhan, 2020, p. 7). Therefore, the only way people will escape unemployment risk is by learning some skills in AI technology.
However, there will be uneven job distribution because not all workers will have access to AI education and training. AI has cost complications, which is an ethical consideration because the learning expenses make it impossible to upskill disadvantaged populations (Shiohira, 2021). In addition, AI technology requires specialized technical skills to design, develop, and implement. This could lead to an uneven distribution of jobs, with workers who possess these skills having greater access to job opportunities and higher salaries, while workers who lack these skills may struggle to find employment or may be limited to lower-paying jobs. Unfortunately, the achievement gap between the rich and the poor has been widening over the last few decades (Sandsør et al., 2023). This indicates that a larger population of the poor is most likely to land low-paying jobs or become jobless.
Consequently, high rates of unemployment result in social disorder. Su (2018) argues that unemployment will cause a significant amount of social unrest before AI’s ethical issues are addressed. Looking at the previous technological advancements, Su (2018) claims during the 60s and 80s period, the rates of divorce and suicide increased significantly in Ohio. This finding suggests that unemployment causes domestic violence and mental health issues. Other researchers argue that it may lead to increased crime rates (Lundqvist, 2018). Therefore, it will be appropriate to ensure equal access to AI training.
Inequality
Inequality is one of the most significant challenges that the world is facing today. As a result, reducing inequality and making sure no one is left behind is one of the UN Sustainable Development Goals (United Nations, 2018). On the other hand, industries are using artificial intelligence to promote sustainable economic development and nurture innovation. However, while AI has the potential to transform various sectors of the economy and improve human well-being, previous research shows that it is also causing significant inequalities in society (Shiohira, 2021). The impact of AI on the labor market is one of the significant factors causing inequality (Korinek & Stiglitz, 2018). As AI can automate a wide range of tasks, it may replace human workers in different industries, resulting in job losses. People who may not have the necessary skills or resources to find new jobs will face difficulties in finding alternative employment, which can worsen existing inequalities between individuals with stable jobs and those without. This postulates that adopting AI in the workplace will increase income inequality because the wage gap between skilled and unskilled laborers will widen.
Another way in which AI is causing inequality is through its potential to spread and amplify biases. AI algorithms rely on large datasets to learn patterns and make predictions. However, if these algorithms are trained on biased datasets, they may perpetuate and amplify those biases, leading to discriminatory outcomes for particular groups (Devillers et al., 2021). For example, there are claims that facial recognition software has lower accuracy rates for people with darker skin tones, which can lead to authorities discriminating against people with dark skin (Najibi, 2020). Another example is a system that Amazon developed to hire people for technical jobs. According to Devillers et al. (2021), the system downgraded all resumes from women, and eventually, Amazon was forced to abandon it after it failed to treat gender terms as neutral. Therefore, biases in AI systems can worsen the issue of inequality instead of improving it.
In addition, AI has the potential to create wealth imbalances that favor a few companies and individuals. In particular, innovators will be positively affected by AI, as their incomes are likely to experience a dramatic increase. Conversely, displaced workers lose their income in the short run due to innovation, worsening the inequality gap between them and innovators (Hadley, 2020). The process of developing and implementing AI technologies might be too expensive for smaller companies, especially those in rural areas (Shiohira, 2021). As a result, a small number of major tech corporations, such as Amazon and Uber, that have already established themselves in the AI sector can further strengthen their competitive edge and market influence (Hadley, 2020). This results in wealth disparities that favor a small number of individuals and companies, worsening existing inequalities.
Furthermore, AI exacerbates inequality levels between developing and developed countries. Luengo-Oroz et al. (2021) argue that artificial intelligence could potentially worsen the inequality in the world instead of tackling it and bringing it to an end. While developed countries have the resources to develop and implement AI systems, developing countries lack the resources to invest in research and development, as well as ICT infrastructure (Shiohira, 2021). Critics may argue that developing countries should implement AI systems to enhance their productivity. However, these nations are already struggling with high unemployment and poverty rates, which leads to low investments in AI systems (Shiohira, 2021). Therefore, AI adoption is likely to heighten inequality levels among countries.
In most cases, high levels of inequality do not last for extended periods because they are often interrupted by human-made catastrophes that result in an equal distribution of resources and opportunities. However, some argue that policymakers can affect distributional outcomes and inequality through public policy (Polacko, 2021). Moreover, Devillers et al. (2021) suggest that AI system developers should be highly aware of the significance of ethical guidelines and the importance of fairness and impartiality, especially with the increase in demand for AI applications. Putting human values at the core of AI can help mitigate the risks associated with AI and inequality.
Data Security and Privacy
Technology has a significant impact on how information is collected, stored, retrieved, and shared. However, its most essential ethical impact is related to the accessibility and manipulation of information (Britz, n.d.). Technology has the potential to increase access to information for more people simultaneously. This means that more people can access a person’s private information, making it easier for the information to be manipulated or used in unethical ways. For instance, AI can be used to automate cyber-attacks that are more sophisticated and challenging to discover. These AI-enabled cyber-attacks can be used to pilfer personal information, financial data, or intellectual property, causing substantial security and privacy concerns. For instance, AI-fueled phishing attacks can masquerade as legitimate individuals or organizations, causing significant financial losses and identity theft (Guembe et al., 2022). As a result, institutions, including hospitals, have to ensure they have a strong security system to protect the data from any unauthorized disclosure (Santosh & Gaur, 2021). Thus, organizations must work hand in hand with AI developers to ensure that their AI systems are programmed and in parallel with the ethical rules to avoid compromising the company’s integrity.
Research also shows that AI can interfere with users’ privacy. Devillers et al. (2021) define privacy as “the right to be left alone” (pg. 78). However, AI can be used to monitor individuals’ movements and behaviors without their consent, which leads to significant privacy concerns. For example, AI-powered surveillance systems can track individuals’ movements and interactions in public spaces, which can be used to identify and track individuals without their knowledge. In addition, another person can use sentiment analysis to scrutinize a worker’s emails and communication on social media handles (Lane & Saint-Martin, 2021). Going through another person’s private life is a violation of privacy, and it nullifies the epistemic privilege. Epistemic privilege refers to when a person knows more about themselves compared to how others know them, and they choose what to reveal and what to conceal (Elliott & Soifer, 2022). This is common in workplaces, especially when you do not know whom to trust with your information.
Furthermore, people use AI to generate deep fakes to spread misinformation, defame individuals, or manipulate public opinion, leading to significant privacy and security concerns. For example, deep fakes could be used to create fake news stories or political propaganda, which can have significant impacts on elections and public policy (Johnson & Johnson, 2023). The idea of portraying something that never happened, in reality, can put the user in a difficult situation.
Some may argue that most of the gadgets that people use to store their information have security settings that can help protect their data. While this claim is true, research shows that most users rarely make an effort to actively protect their data. In fact, most of them give it away voluntarily without knowing (Devillers et al., 2021). For instance, when logging in to apps and websites, most people tick the box without even reading the terms and conditions. However, some conditions might inform the user that their data will be used by third parties. Thus, blindly ticking the box or accepting cookies can expose the users’ data to cyber criminals unknowingly.
Potential Benefits
While it is true that AI has raised ethical concerns, its numerous benefits in various industries cannot go unnoticed. First, AI can analyze large amounts of data quickly and accurately, providing insights that can inform better decision-making (Lane & Saint-Martin, 2021). This can be particularly valuable to organizations such as healthcare, where AI can assist in diagnosis and treatment decisions. Second, AI increases efficiency and productivity, considering that systems powered by AI complete tasks faster and with more accuracy. One of the major fields that have embraced AI in health facilities. AI can assist in diagnosis, treatment planning, and drug discovery, improving the overall efficiency of healthcare delivery. Additionally, AI-powered medical devices can provide more accurate and reliable results, improving patient outcomes (Nadimpalli, 2017). Therefore, it saves the time that physicians would spend consulting on health issues that patients could be experiencing.
Artificial intelligence is also used by pharmaceutical companies to determine the different characteristics of drugs and their potential side effects (Nadimpalli, 2017). This increases productivity since AI takes lesser time compared to one that pharmacists would take to outline the side effects. “In 2020, the AI program AlphaFold predicted a protein’s 3D structure and successfully applied AI to solve a complex grand challenge that biologists had been working on for over fifty years” (Shiohira, 2021, p. 11). These advancements have fueled fears about job loss, seeing that AI is causing not only higher productivity but also improved job quality (Lane & Saint-Martin, 2021). Economists argue that AI will create jobs, while other people claim that it will destroy or replace some of the existing jobs. However, ethical concerns must be addressed to ensure everyone in society enjoys these benefits.
Conclusion
The adoption of AI technology has raised ethical concerns, particularly regarding unemployment, privacy, and inequality. While some argue that AI will create more jobs, history shows that the widespread adoption of technology causes significant unemployment, particularly among medium-skilled workers. Additionally, the cost of AI education and training and the need for specialized technical skills may lead to an uneven distribution of jobs, with the poor and disadvantaged populations struggling to find employment or being limited to lower-paying jobs. Moreover, the impact of AI on the labor market and its potential to amplify biases can lead to significant inequalities in society. It is, therefore, essential to address these ethical concerns to ensure that the benefits of AI technology are shared equitably and that no one is left behind. This can be achieved by ensuring equal access to AI training, addressing biases in AI algorithms, and promoting policies that prioritize the well-being of all individuals.
References
Bordot, F. (2022). Artificial intelligence, robots, and unemployment: Evidence from OECD countries. Journal of Innovation Economics & Management, 1(37), 117–138. DOI: 10.3917/jie.0037.0117
Britz, J.J. (n.d.). Technology as a threat to privacy: Ethical challenges to the information profession. The University of Pretoria. Retrieved from http://web.simmons.edu/~chen/nit/NIT’96/96-025-Britz.html
Devillers, L., Fogelman-Soulie, F., & Baeza-Yates, R. (2021). AI & human values: Inequalities, biases, fairness, nudge, and feedback loops. In B. Braunschewig & M. Ghallab (Eds.). Reflections on artificial intelligence for humanity. Springer Nature
Elliott, D. & Soifer, E. (2022). AI Technologies, Privacy, and Security. Frontiers in Artificial Intelligence, 5(826737), 1–8. doi: 10.3389/frai.2022.826737
Guembe, B., Azeta, A., Misra S., Osamor, V.C., Fernandez-Sanz, L. & Pospelova, V. (2022). The emerging threat of AI-driven cyber-attacks: A review. Applied Artificial Intelligence, 36(1), 2376-2409. https://doi.org/10.1080/08839514.2022.2037254
Hadley, J. (2020). Artificial intelligence and rising inequality. The Undergraduate Research Writing Conference, Rutgers, 1–16. https://sites.rutgers.edu/nb-senior-exhibits/wp-content/uploads/sites/442/2020/08/James-Hadley-final-pdf.pdf
Johnson, D. & Johnson, A. (2023). What are deepfakes? How fake AI-powered media can warp our perception of reality. Insider. Retrieved from https://www.businessinsider.com/guides/tech/what-is-deepfake?r=US&IR=T
Korinek, A. & Stiglitz, J. (2018). Artificial intelligence and its implications for income distribution and unemployment. In A, Agrawal, J. Gans, & A, Goldfarb, The economies of artificial intelligence: An agenda (pp. 349-390). National Bureau of Economic Research.
Lane, M. & Saint-Martin, A. (2021). The impact of artificial intelligence on the labor market: What do we know so far? OECD Social, Employment and Migration Working Papers No. 256, 1–60. https://dx.doi.org/10.1787/7c895724-en
Lee, H. J., Probst, T. M., Bazzoli, A., & Lee, S. (2022). Technology advancements and employees’ qualitative job insecurity in the Republic of Korea: Does training help? employer-provided vs. self-paid training. International journal of environmental research and public health, 19(21), 14368. https://doi.org/10.3390/ijerph192114368
Liu, R. & Zhan, Y. (2020). The impact of artificial intelligence on job insecurity: A moderating role based on vocational learning capabilities. Journal of Physics: Conference Series, 1629(012034), 1–8. doi:10.1088/1742-6596/1629/1/012034
Luengo-Oroz, M., Bullock, J., Pham, K. H., Lam, C. S. N., & Luccioni, A. (2021). From artificial intelligence bias to inequality in the time of covid-19. Ieee Technology and Society Magazine, 40(1). https://doi.org/10.1109/MTS.2021.3056282
Lundqvist, F. (2018). Unemployment and crime. Södertörns högskola, 1-19. http://www.diva-portal.org/smash/get/diva2:1252037/FULLTEXT01.pdf
Nadimpalli, M. (2017). Artificial Intelligence Risks and Benefits. International Journal of Innovative Research in Science, Engineering, and Technology, 6(6). https://www.researchgate.net/publication/319321806_Artificial_Intelligence_Risks_and_Benefits
Najibi, A. (2020). Racial discrimination in face recognition technology. Science Policy and Social Justice. Retrieved from https://sitn.hms.harvard.edu/flash/2020/racial-discrimination-in-face-recognition-technology/
Palacko, M. (2021). Causes and consequences of income inequality – An overview. De Gruyter, 12(2), 341-357. https://doi.org/10.1515/spp-2021-0017
Sandsør, A. M. J., Zachrisson, H. D., Karoly, L. A., & Dearing, E. (2023). The widening achievement gap between rich and poor in a Nordic Country. Educational Researcher, 0(0). https://doi.org/10.3102/0013189X221142596
Santosh, K., & Gaur, L. (2021). Privacy, security, and ethical issues. In: Artificial intelligence and machine learning in public healthcare. SpringerBriefs in Applied Sciences and Technology (pp. 65–74). https://doi.org/10.1007/978-981-16-6768-8_8
Shiohira, K. (2021). Understanding the impact of artificial intelligence on skills development. UNESCO and UNESCO-UNEVOC International Centre for Technical and Vocational Education and Training, 1–55. https://files.eric.ed.gov/fulltext/ED612439.pdf
Su, G. (2018). Unemployment in the AI age. AI Matters, 3(4), 35–43. https://doi.org/10.1145/3175502.3175511
United Nations (2018). Leaving no one behind. United Nations CDP Committee for Development Policy. Retrieved from https://sustainabledevelopment.un.org/content/documents/2754713_July_PM_2._Leaving_no_one_behind_Summary_from_UN_Committee_for_Development_Policy.pdf
write