{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# 1. Introduction\n\nArtificial intelligence and machine learning are powerful tools that are rapidly evolving and being integrated into our daily lives. Notably, generative AI such as ChatGPT and backend machine learning algorithms in mobile applications like TikTok have revolutionized the way we create and consume content. Naturally, these tools have found a plethora of uses in the workplace, and this essay will outline the latest studies on how artificial intelligence is changing the workplace, and the ethical implications of AI use in the workplace.","metadata":{}},{"cell_type":"markdown","source":"# 2. The Recruitment Process\n\nIn the beginning of every career is a recruitment process one must undergo to qualify for a position in a company, but how does artificial intelligence affect recruitment today?\n\nArtificial intelligence (AI) has revolutionized recruitment by providing companies seeking new employees with remarkable effectiveness and efficiency. Recruiters may refine their methods and expedite the entire process with the help of AI, transforming different elements of recruitment. The use of advanced algorithms by AI in resume screening allows for quick and complete review of applicants. When compared to human recruiters, AI can swiftly read and analyze a big number of resumes, saving a lot of time. The most qualified individuals may be found through this fast screening method, which takes into account factors including education, work experience, and talents.\n\nAI's usage in hiring also results in measurable improvements in accuracy and efficiency. AI shortens the time and effort needed for preliminary screening by quickly scanning resumes and selecting the best prospects. This streamlined procedure guarantees that the best candidates are quickly discovered, improving recruitment results. Also, using AI for resume screening encourages impartiality and reduces bias. AI enables a more objective screening process and equal possibilities for all candidates thanks to its sophisticated algorithms. This aids businesses in creating a more inclusive and diverse pool of candidates, which eventually results in better hiring decisions.\n\n\n## 2.1 Collaboration Among Recruiters and Artificial Intelligence: Removing Human Prejudices in Employment\n\nThis study by Zhisheng Chen [1] outlines suggestions on how to incorporate AI in the recruitment process to create a more diverse and inclusive workplace without the prejudice of human recruiters. Chen explains that talent acquisition is now in a digital 3.0 environment where AI is given the power to scan candidates and make decisions on who to hire. Chen suggests a deep learning method that factors in candidate experience, education, skills, and duties, feeds it into a two-layer neural network that outputs an error and backpropogates to reduce error in the hopes of learning about the candidates in the most accurate way possible. Furthermore, Chen emphasizes human-AI interaction, leading to the decision on who to hire to remain in the hands of humans who analyze the AI model's output critically.\n\n## 2.2 Ethical Implications\nIn theory, using AI to aid in the recruitment process can help ensure that prejudice on the basis of identity is minimized as humans, with their innate prejudices, are imperfect and will have their biases towards candidates on a non-merit basis. However, it is important to keep in mind that AI is trained by humans, and the biases of humans will still seep into the AI on the training data it is given, and the final decisions made by the humans from the AI's output. It is therefore important to be cautious of using AI for talent acquisition because while an AI algorithm can try to be as unbiased as possible, there is an inevitable prejudice that will show up since humans create the AI.\n\nThe question here is how we might address the ethical concerns around user privacy and AI bias. As an AI system developer constructs the AI, they should be aware of the legal requirements for data collecting and avoid breaking the law. Nevertheless, when laws are imposed or people severely limit their data on media, the performance of the developers' AI systems suffers because they typically require a large amount of data to be trained on. To properly employ AI-powered recruitment procedures, businesses must invest in individuals who are well-versed in the risks and potential associated. This involves hiring professionals and data scientists who are aware of and sensitive to any biases and weaknesses in the algorithms. It is critical to ensure a broad and unbiased training dataset, as well as thorough algorithm design. Frequent audits and checks aid in identifying and correcting any unexpected biases, ensuring fairness and ethical integrity in the recruitment process.\n\nTo summarise, successful AI-powered recruitment implementation necessitates investing in knowledgeable personnel who understand the risks and potential connected with this technology. This includes identifying potential biases and ethical issues, guaranteeing diverse and unbiased training datasets, and designing mindful algorithms. Companies may enhance their recruitment methods while maintaining fairness and ethical integrity by creating a balance between AI technology and human knowledge.\n","metadata":{}},{"cell_type":"markdown","source":"# 3 The State of the Job Market\n\nArtificial intelligence has drastically changed the work landscape. It seems inevitable that the use of AI and ML will be needed for almost all forms of office and service-based employment. So, what have been the latest examples of this changing landscape?\n\n## 3.1 Employment 5.0: The Work of the Future and the Future of Work\n\nThis paper by Kolade and Owoseni [2] suggests the idea of the Fourth Industrial Revolution that is described by the integration of AI, robotics, and big data in the workplace. This has created a vast array of new jobs in the data science and AI fields, but has displaced more traditional, manual jobs that were originally thought to be stable and unchanging. It is now imperative upon new workers that they learn about data science, artificial intelligence, and robotics and they must keep up to date with the latest advancements in technology if they are to succeed in today's job market.\n\nIn Employment 5.0, there are three theoretical perspectives: **Socio-technical system theory**, **Skill-biased technological change** and **Political economy of automation and digital transformation**\n\n
\"Employment
\n\n\n**Social-technical system theory** is a six-dimensional structure encompassing the following components: People, Infrastructure, Technology, Culture, Process/Procedures, and Goals. It offers a holistic perspective on how these elements interact within a complex organizational system. This framework is an effective tool for analyzing technological innovations as a structure comprising interrelated components. It helps to capture the dynamics of human-machine interactions in the Industry 4.0 system. Moreover, it highlights the consequences of these interactions on evolving tasks and the demand for new skill profiles. Therefore, this framework would equip policymakers with the appropriate information to anticipate and address employment challenges and opportunities arising from the ongoing digital transformation.\n\n\nThe '**Skill-biased technological change** 'theorectical perspective aims to explain how the rapid integration of computer technology impacts both the tasks done by workers and the demand for specific human skills. This framework is based on two beliefs: First, computers tend to replace workers in routine tasks that can be easily programmed into computer systems, and second, tasks that lack explicit rules and demand greater flexibility and creativity (non-routine tasks) are less prone to computer substitution. Another skill-biased classification model categorizes tasks into three distinct types: perception and manipulation tasks, creative intelligence tasks, and social intelligence tasks\n\n- Perception and manipulation tasks are tasks that require finger and manual dexterity and tasks that involve working in cramped and awkward spaces\n- Creative intelligence tasks encompass the ability to generate original and unconventional ideas for problem-solving, along with proficiency in fine arts such as music, dance, and visual arts.\n- Social intelligence tasks involve skills like social perceptiveness, negotiation, and persuasion.\n\nWhile perception and manipulation tasks are highly susceptible to substitution, tasks related to creative and social intelligence face a lower risk of being replaced, primarily because they are non-routine and less amenable to computer coding. Nevertheless, none of these task categories are entirely immune to substitution.\n\nThe ‘**Political economy of automation and digital transformation**’ theoretical perspective discussed several scholars that approached automation and digital transformation from a critical theory perspective and criticize existing frameworks used to study the impact and future of digital transformation. Their argument centres on the idea that digital technologies are not neutral but instead shaped by unequal power dynamics, favouring capital over labour. Consequently, these technologies prioritize increasing surplus value, often at the expense of human workers, potentially leading to job displacement and reduced wages. They propose an alternative framework that highlights the importance of shared or collective ownership of technology. They view technology as a positive force that, when owned collectively, can empower workers rather than subjugate them.\n\n\n## 3.2 Ethical Implications\n\nWith the Fourth Industrial Revolution comes consequences for unemployment and job displacement. With many easily automatable jobs being replaced my AI, it raises concerns that we are beginning to rely too heavily on artificial intelligence to automate many of our jobs. As a result, it becomes more difficult for those with less access to resources to learn about the latest technological advancements to move up the corporate ladder and to even find a job at all. It narrows the scope of the types of jobs we will have in the future. Regulating how much we depend on AI can ensure that we not only make sure not too many people are losing their jobs, but it also gives less power to the companies that invent the AI tools that replace those jobs in the first place.\n\n## 3.3 Job Attitudes and Career Behaviors Relating to Employees’ Perceived Incorporation of Artificial Intelligence in the Workplace: a Career Self-Management Perspective\n\nThis article by Presbitero and Teng-Calleja [3] conducts a survey about the perceptions of employees on how artificial intelligence is being integrated into their jobs and how they feel about the security of their job as a result. It gives insight into the general job market sentiment in 2023 and the perceived direction it is going in for the future. \n\nWhen individuals make career choices, personal experiences play a significant role. It is believed that employees' **sense of uncertainty** and **feelings of job insecurity** are influenced by their **workplace experiences** (e.g. feeling threatened by AI integration), which can impact their career exploration behaviour. One can argue that the behaviour of actively searching for **alternative career opportunities** is driven by the expectation that other careers might offer better conditions, leading to reduced psychological distress\n\nFive hypotheses were explored:\n- Employees’ perception of AI taking over jobs is positively and significantly associated with their career exploration behaviour.\n- Perception of AI taking over jobs is positively and significantly associated with job insecurity.\n- Job insecurity mediates the relationship between the perception of AI taking over jobs and career exploration behaviour.\n- Job insecurity is positively and significantly associated with psychological distress.\n- Psychological distress mediates the relationship between job insecurity and career exploration behavior.\n\nIt was discovered that as AI continues to be deployed and integrated into organizations, it poses a potential threat to the job security of many employees. When employees experience high levels of physical, emotional, or psychological distress in the workplace, they are more likely to engage in **adaptive career behaviours**, such as career exploration. It was also found that heightened experience of psychological distress due to job insecurity could influence an employee’s career exploration behaviour.\n\n## 3.4 Ethical Implications\nWhen the job market feels a certain way, it influences how available it is for employers to recruit from. When everyone does not know whether they will keep their job, they will either want to hold on to it and not let go, or will leave preemptively and try to develop themselves to find a more secure job. This shift will have reprecussions on the availablity of employees for jobs that are not yet automated but may be in the future.","metadata":{}},{"cell_type":"markdown","source":"# 4. Optimizing Workplace Productivity and Outcomes\n\nNaturally, since AI has proven to be a powerful tool in a wide variety of situations, its applications to boosting work productivity is an active field of research whose results employers seek to implement in their workplaces. Finding a way to boost employee productivity accelerates a company's growth and progress, leading to larger profits and a larger return on investment in paying employees.\n\n## 4.1 Machine learning and role of artificial intelligence in optimizing work performance and employee behavior\n\nThis article by Ramachandran et al. [4] details the capabilities of artificial intelligence and machine learning in automating repetitive tasks in the workplace so that employees can invest more time in performing tasks that require critical thinking and human application of their knowledge. It argues that businesses can make more money when employees are more satisfied with their work when repetitive task completion is accelerated by AI, and the accuracy by which AI performs these tasks exceeds that of humans, leading the authors to suggest that employees and customers alike would be happier when AI is delegated more tasks.\n\n## 4.2 Ethical Implications\nOne must be careful not to rely too heavily on AI for some of these tasks. What consistutes a \"repetitive task\" may be important for a human to manually review before allowing AI to perform it, or may require a human to complete it themselves. For example, working with sensitive patient data in a hospital needs oversight from humans when it comes to making decisions based on that data, and its organization needs to be supervised by humans to ensure it is properly being stored.\n\n## 4.3 How Artificial Intelligence Technology Affects Productivity and Employment: Firm-level Evidence from Taiwan\nThis paper by Chih-Hai Yang [5] conducts an empirical analysis on how the country of Taiwan is using AI to boost employee productivity. It concludes that while AI is able to increase productivity significantly, it has its negative impacts on employment, as many jobs will be displaced or replaced by AI entirely, leading to a growing worry of a rising unemployment rate in the country.\n\n## 4.4 Ethical Implications\nWhen many jobs are automated, this can produce much more accurate results and generally will positively impact the output of a company. However, this comes at the cost of a higher unemployment rate for a given country, which may inadvertently slow down the economy due to a slower exchange of goods and services when people do not have a steady source of income to spend from, leading to a cycle of more automation leading to less jobs leading to less spending from the public and less profit for the companies. This is, in fact, the exact opposite result of the perceived increase in company profits for many businesses.\n\n## 4.5 A Systematic Literature Review on the Impact of Artificial Intelligence on Workplace Outcomes: A Multi-Process Perspective\nThis literature review by Pereira et al. [6] aims to summarize how AI is being used to increase workplace outcomes. It concludes that while AI has a positive impact on productivity, job satisfaction, and customer satisfaction, it also leads to larger economic inequality for individuals whose jobs become displaced by AI, and are therefore required to adapt to the new job market.\n\n## 4.6 Ethical Implications\nThe demographics of those who are most affected by the changing job market and job displacement are either part of an older generation of employees who have not been able to keep up with the ever-evolving world of AI, or those with less formal education, who may not have access to those institutions to learn about the latest advancements in AI. These employees would therefore struggle to remain in the workforce, which would require them to adapt. However, when too many jobs are displaced at once, it can cause instability in the job market. Therefore, it is important to apply AI into the workplace in a gradual fashion, or have humans work alongside AI to prevent this sudden instability.","metadata":{}},{"cell_type":"markdown","source":"# 5. AI and Employee Creativity\n\nCorporations are increasingly investigating the use of AI to take on more mundane and repetive tasks. This section will analyze how this will affect employee creativity in the workplace\n\n## 5.1 When and How Artificial Intelligence Augments Employee Creativity\n\nThis article by Nan Jia et al. [7] studies AI and human collaboration in a sequential division of labor (assigning well-codified and repetitive tasks to AI and assigning the higher-level portions, which require more problem-solving, to human employees). In the study conducted, AI assistance in generating sales leads led to an increase in employee creativity in developing more persuasive sales pitches, as evidenced by higher sales. However, the effect was significantly more evident in employees with a high skill level, since they were able to develop more creative sales pitches, given the basic AI generated one. In contrast, lower-skilled employees were not able to contribute much to add onto AI generated scripts and hence experienced negative emotions at work. Thus the article concludes that the sequential division of labor can result in greater employee creativity, but primarily for employees with greater job skills.\n\n## 5.2 Ethical Implications\n\nWhile the use of AI to take on highly-codified and repetitive tasks can be used to tangibly increase employee creativity, this is heavily dependent on the employee skill level. This means that employees who may perform well in well-defined, yet less creative tasks, may be at risk of being replaced with AI in the future. This may increase the inequality gap as those who do not have access to higher education may be more limited in experiences (e.g. higher-skill level jobs) or learning (e.g. the variety of tasks and case studies examined in universities to develop a broad range of skills) pertinent to their employment and may thus have a more difficult time contributing to the work AI has produced. Furthermore, the use of AI in completing more codified and repetitive tasks has been seen to have a negative impact on the workplace satisfaction of lower-skilled employees. Thus it is important for corporations to comprehensively evaluate whether they can implement AI in a way that limits the negative implications mentioned above. \n\n## 5.3 Influence of Artificial Intelligence and Robotics Awareness on Employee Creativity in the Hotel Industry\n\nThis study by Hui Wang and et al. [8] explores the influence of AI and robotics awareness (AIRA) on employee creativity in the hotel industry. The authors found that when employees work in a hotel which has AIRA, they will take proactive behaviors to achieve person-environment fit. For example, it was found that employees engage in more active learning when working with AI and robotics. This is beneficial because active learning leads employees to enrich their skills and knowledge which enhances their creativity. Another example of proactive behavior includes task crafting, which is when employees alter the extent or number of tasks in their job to align better with their skills and interests. This allows employees to cooperate better with AI and robotics and hence benefits employee creativity. Task crafting also allows employees to generate various ideas and examine their tasks from unconventional perspectives thus promoting creative thinking.\n\n## 5.4 Ethical Implications\n\nThe promotion of active behaviors such as active learning and task crafting not only benefits employee creativity, but also helps employees to cooperate productively with AI. For example, active learning helps employees gain the knowledge and skills to coexist productively with AI, while task crafting helps employees reshape their job to eliminate or reduce overlap with the work that AI can produce, hence allowing them to focus on the parts that AI have limited capacity to do i.e. the less codified and repetitive parts of a job which often bring greater employee satisfaction. However, this is only possible for employees that are willing and able to take on the challenge of developing these active behaviors. This hence puts employees who are not able to put in this energy, whether for personal, health, or other reasons, at risk. Thus when implementing AI, corporations should prioritize the cooperation between employees and AI. That is, the focus should be on training the employees to learn to use assistance from AI to effectively increase their workplace creativity, rather than seeing AI as competition for their jobs.\n","metadata":{}},{"cell_type":"markdown","source":"# 6. Conclusion\n\nIn conclusion, the integration of AI tools into the workplace has many potential benefits. For example, AI can help improve the accuracy of the candidate profiles during the recruitment process,and the delegation of mundane, repetitive tasks to AI can improve both workplace productivity and creativity. However, it is also important to consider the many ethical considerations of the incorporation of AI into the workplace. For instance, since AI are trained by humans, they are subject to human biases which may be detrimental to some demographics of candidates during the recruitment process, and human employees may feel their livelihoods are threatened by AI which may reduce productivity. It may also limit the ability of employees with lower skill to contribute and hence result in demotivation. Altogether, the integration of AI into the workplace can have many positive effects, but it is important to consider the ethical implications of doing so.\n","metadata":{}},{"cell_type":"markdown","source":"# References\n[1] Z. Chen, “Collaboration among recruiters and artificial intelligence: removing human prejudices in employment,” vol. 25, no. 1, pp. 135–149, Feb. 2023, doi: 10.1007/s10111-022-00716-0. [Online]. Available: https://link.springer.com/article/10.1007/s10111-022-00716-0\n\n[2] O. Kolade and A. Owoseni, “Employment 5.0: The work of the future and the future of work,” vol. 71, p. 102086, Nov. 2022, doi: 10.1016/j.techsoc.2022.102086. [Online]. Available: https://dx.doi.org/10.1016/j.techsoc.2022.102086\n\n[3] A. Presbitero and M. Teng-Calleja, “Job attitudes and career behaviors relating to employees’ perceived incorporation of artificial intelligence in the workplace: a career self-management perspective,” vol. 52, no. 4, pp. 1169–1187, May 2023, doi: 10.1108/PR-02-2021-0103. [Online]. Available: https://www.emerald.com/insight/content/doi/10.1108/PR-02-2021-0103/full/html\n\n[4] K. K. Ramachandran, A. Apsara Saleth Mary, S. Hawladar, D. Asokk, B. Bhaskar, and J. R. Pitroda, “Machine learning and role of artificial intelligence in optimizing work performance and employee behavior,” vol. 51, pp. 2327–2331, 2022, doi: 10.1016/j.matpr.2021.11.544. [Online]. Available: https://dx.doi.org/10.1016/j.matpr.2021.11.544\n\n[5] C.-H. Yang, “How Artificial Intelligence Technology Affects Productivity and Employment: Firm-level Evidence from Taiwan,” vol. 51, no. 6, p. 104536, Jul. 2022, doi: 10.1016/j.respol.2022.104536. [Online]. Available: https://dx.doi.org/10.1016/j.respol.2022.104536\n\n[6] V. Pereira, E. Hadjielias, M. Christofi, and D. Vrontis, “A systematic literature review on the impact of artificial intelligence on workplace outcomes: A multi-process perspective,” vol. 33, no. 1, p. 100857, Mar. 2023, doi: 10.1016/j.hrmr.2021.100857. [Online]. Available: https://dx.doi.org/10.1016/j.hrmr.2021.100857\n\n[7] N. Jia, X. Luo, Z. Fang, and C. Liao, “When and how Artificial Intelligence Augments Employee creativity,” Academy of Management Journal, 2023. doi:10.5465/amj.2022.0426\n\n[8] H. Wang, H. Zhang, Z. Chen, J. Zhu, and Y. Zhang, “Influence of Artificial Intelligence and robotics awareness on employee creativity in the hotel industry,” Frontiers in Psychology, vol. 13, 2022. doi:10.3389/fpsyg.2022.834160 \n","metadata":{}},{"cell_type":"code","source":"import pandas as pd\n\nsubmission = {\"type\": [\"essay_category\", \"essay_url\", \"feedback1_url\", \"feedback2_url\", \"feedback3_url\"],\n \"value\": [\"Other\", \"https://www.kaggle.com/code/rosyiqdaniel/ai-and-the-future-of-work\", \"https://www.kaggle.com/code/jbrans/ethics-in-the-ai-wonderland/comments#2339834\", \"https://www.kaggle.com/code/lorresprz/applying-ai-ml-to-theoretical-physics/comments#2330935\", \"https://www.kaggle.com/code/pluvias/kaggle-ai-report-optimization/comments#2330947\"]}\nsubmission = pd.DataFrame(submission)\nsubmission.to_csv(\"/kaggle/working/submission.csv\", index=False)","metadata":{"execution":{"iopub.status.busy":"2023-07-16T11:55:35.726681Z","iopub.execute_input":"2023-07-16T11:55:35.727218Z","iopub.status.idle":"2023-07-16T11:55:35.780579Z","shell.execute_reply.started":"2023-07-16T11:55:35.727163Z","shell.execute_reply":"2023-07-16T11:55:35.779555Z"},"trusted":true},"execution_count":null,"outputs":[]}]}