The Conflict of Artificial Intelligence Technology in Employment

Analysis

In Finley's article, he highlights five of the best practices employers can follow in order to manage the risk when using AI tools: (1) knowing the data; (2) disclosing the topics and methodology; (3) considering undergoing a bias audit; (4) implementing human oversight; (5) and reviewing vendor agreements. It was also stated that the NLRB general counsel proposed a plan to "crack down on electronic monitoring in the workplace based upon concerns of infringement upon employees' rights to engage in protected concerted activity" (Finley, 23). This new framework could possibly reduce the growth of "smart" workplaces across America, if successful. Technically, the goal is to enforce restrictions on electronic monitoring in order to protect the rights of employees to partake in such activities within or outside the workplace. Farrell highlights one key disadvantage of AI on the job. It is said that "potential bias in algorithmic decision-making, rooted in the adaptivity and autonomy of AI, may result in discriminatory outcomes" (Farrell, 54). In this case, there may have been concerns over possible violations of both the Equality Act 2010 and Human Rights Act 1998 stemming from such bias in the AI software. Kim and Bodie also criticized the functionality of artificial intelligence in employment settings. For instance, "if an employer is using a biased algorithm because it wants to screen out members of a protected group, that is clearly a form of intentional discrimination prohibited under disparate treatment theory" (Kim & Bodie). Disparate treatment theory is a theory that bans unfavorable decisions that are made based on race, color, sex, religion, or other protected class. Belusic and Samardzija argues that workers need to adapt to certain changes when it comes to AI being used in the workplace. When artificial intelligence started taking over mechanical jobs, people become focused more on thinking jobs. According to the authors, now when artificial intelligence is successfully taking over thinking jobs, "people tend to find feeling jobs as artificial intelligence is still not developed enough to complete these" (Belusic & Samardzija, 84). Research has shown that certain jobs would focus on feelings and that workers would be "people-oriented" and less data-oriented. Chen Guang conducted research on AI technology in construction jobs. In terms of safety on the job, physical and physical insecurity and their interaction are the direct causes of security incidents, all while the number of temporary structures and facilities are big, such conditions on the ground have significantly changed. On top of that, physical insecurity (working environment) takes place throughout the system. To a large extent, "construction insecurity is also caused by the insecurity of people, especially managers, such as unsafe construction schemes, unreasonable safety management methods or safety measures" (Guang, 6631). Basically, the purpose of incorporating AI technology in the construction industry is to improve safety levels for its workers. Frances Flanagan and Michael Walker addresses the problem with employees giving sensitive data to sources like Facebook and/or third parties. Far from being worker-owned/controlled, "information on Facebook was subjected to surveillance, advertising and was non-portable and vulnerable to being shut down or modified as a consequence of unilateral changes of rules and/or functionality" (Flanagan & Walker, 164). This is a sign that employees should be very careful on where they give their information, because such information should never be allowed on unsecure sources. Sonal Pandey addresses the perspective of individuals who are pessimistic that there still will be jobs for human workers, while automation is growing. According to Pandey, "the higher the organizational degree and focus on workforce and technology collaborations, the higher the productivity would be" (22). For instance, production has increased by 85% due to great collaboration of human workers and machines in the BMW factory, according to the MGI Report. 

Lilla Vicsek discusses the analytical component of fictional expectations, which is agency. According to her argument, "agency is awarded to AI technology from both a Positive Effects and Negative Effects perspective, whilst reduced agency is often only attributed to humans, governments, or other organizations (in relation to getting ready for responding to changes caused by technology)" (Vicsek, 8). The author is pointing out the difference between the concept of agency in relation to artificial intelligence and the type of agency connected to people, the government, or any other entity. Charlotte Stix addresses the meaning behind the term "trustworthy AI." It was stated that: "In order for an AI system to count as "trustworthy," (1) it must be lawful; that is, adhering to all legal obligations which are binding and required at that time, (2) it should be ethical; that is, adhering to and fulfilling all ethical key requirements that have been put forward in the Ethics Guidelines for Trustworthy AI, and (3) it should be robust, both from a technical and a social perspective" (Stix). The so-called "trustworthy" AI is defined as being composed of three parts in its final form. Lastly, research has shown descriptive statistics and correlations regarding AI. Researchers reported that "AI application was not significantly correlated with the perception of unemployment risk" (Xu & Xue, 5). The reason for this result might have been due to the fact that AI application comes with both positive and negative effects on employees. However, certain risks leading to unemployment should never be taken lightly.