The current healthcare state is defined by the complexity and rise of data, which has necessitated the introduction of artificial intelligence (AI). In many healthcare facilities, healthcare administrators are tasked with roles ranging from ensuring better patient experiences, patient record keeping and other responsibilities such as designing budgets. Therefore, it is imperative to integrate AI components, such as machine learning, in policy and decision-making among different administrative roles (Lee, 2022). The following executive summary centralizes the application of AI in healthcare administration.
Background
Healthcare administration or non-patient care tasks entail all the work performed behind the scenes to ensure patients gain better healthcare experiences. Additionally, the administrative process deals directly with budgets and policies, ensuring the healthcare providers’ welfare is met, and their safety is warranted (Lee, 2022). Some of the key responsibilities this arm of healthcare is tasked with include developing staff and physician schedules, managing finances, improving facility quality and efficiency, training staff and ensuring law compliance.
Gap Description
Healthcare Issue Description
As described earlier, the over-reliance of people without appropriate technology in executing healthcare administrative roles often translates to bureaucracy, resulting in hospital inefficiencies. With hospital inefficiencies, the organization’s workflow is interfered with, leading to huge hospital losses. According to Forehand et al. (2021), about $1.75 million per hospital is lost in the US due to inadequate communication between hospital administrators. This figure translates to $11 billion in losses in the US healthcare industry (Forehand et al., 2021). Besides ineffective communication, most hospitals lacking AI systems experience inefficiencies such as document duplication, poor patient flow, inappropriate hospital admissions and length of stay and incomplete drug reconciliation.
Stakeholder Groups Impacted by the Issue
The main stakeholder groups affected by healthcare inefficacies from the organization administration are patients, physicians and nurses. Healthcare inefficiency affects patients’ flow from one department to another or even in and out of the hospital. Most patients succumb after undergoing long waiting periods for their surgeries and treatments. Besides, physicians face the challenge of document duplication in a case where there are inefficiencies in the documentation methods (Lee, 2022). For instance, physicians might find duplicated charts or electronic records, which means much time might have been wasted on a patient that required minor treatment.
Responsible Stakeholder in Addressing the Issue
The key stakeholders in addressing healthcare inefficiencies in the administrative processes include the government, hospital administrators and the direct-patient contact staff. The government should fund technological projects such as robotics and neural work, which aim to reduce hospital administration’s bureaucracy and complexity (Lee, 2022). Besides, hospital administrators should redesign system-wide processes that streamline patient flow in the organization. Lastly, direct-contact staff, such as nurses, should be familiar with communication methods such as texting and medical records.
Research Related to the Gap and New Service Proposal
Healthcare inefficiency can translate to huge losses, meaning prompt strategies should be implemented to counter it. It is estimated that inadequate communication resulting from a lack of secure text messaging in healthcare can lead to a loss of $875,000 per hospital (Mintz & Brodie, 2019). Therefore, it is important to introduce AI in a hospital setting to offset the present inefficacy. AI has five subsets that are imperative to healthcare administration: neural network, machine learning, deep learning, robotics and computer vision. This technology can improve communication by catching subtle cues to precisely analyze the presentation methods. The implementation of the AI technology has been outlined in Table 1.
Detailed Evaluation Plan
Table 1. Evaluation Plan for the Implementation of AI in Healthcare Administration
Strategic Objectives
Key Initiatives (1-to-3-year initiatives)
Integrate machine learning in voice-to-text transcription
Partner with companies such as RapidMiner to understand machine learning
Introduce neural network in identifying patient records
Use Python libraries such as NumPy to acknowledge the working of neural network
Introduce computer vision in gathering important data from patient records
Establish a workgroup that to learn about computer vision through tools such as OpenCV
What is Artificial intelligence? And has it benefited people through its many fascinations? Artificial intelligence has become particularly widespread in the modern world, but there are significant controversies in people’s lives. Hence, this technology has a negative impact on society and causes its significant decline and the emergence of several drawbacks. However, many people are of the opinion that artificial intelligence has multiple positive qualities to improve people’s lives. Despite this, the outweighing disadvantages of artificial intelligence are a rise in unemployment and an increase in laziness.
Discussion
No job satisfaction originates when people start losing jobs they love to artificial intelligence, thus leading them to have jobs they don’t enjoy. Thus, innovative technology covers an increasing number of professions that can be performed without the participation of human resources. Research shows that “unemployment would result because workers could not survive working for the market-clearing wage, and it would pay employers to raise real wages above the level because of the increase in worker productivity” (Korinek & Stiglitz, 2018, p. 352). Moreover, it is pointed out that “jobs with high exposure to automation technologies experienced a decline in employment and wages” (Bordot, 2022, p. 119). Therefore, people might find it hard to find jobs that are good which leads to unemployment which further leads to poverty.
Another significant negative aspect of using artificial intelligence is the lack of inspiration and creativity. These characteristics are an integral part of the work of human consciousness, which cannot be introduced into the process of work using this technology. In addition, the introduction of artificial intelligence has a negative effect on reducing the level of work ethic and enthusiasm. People lose interest in performing the actions assigned to them, as they consider it unnecessary to perform activities that can be replaced by a machine. The lack of initiative also results in obesity and other health issues that arise due to the inactivity of employees. Thus, depending on the technology, it could lead to humans using less knowledge and their brain functions to perform actions.
However, despite all the disadvantages, there is a point of view that artificial intelligence provides significant benefits. Henceforth, it is considered that it has an unlimited time limit. Thus, if people tend to be tired and, in some cases, burnout, innovative technology can work constantly. Moreover, artificial intelligence is characterized by greater efficiency and accuracy. Korinek and Stiglitz (2018) point out that some suggest that “artificial intelligence will mainly assist humans in being more productive, and refer to such new technologies as intelligence- assisting innovation” (p. 350). Moreover, research has shown that the use of new technology contributes to increased productivity, thereby influencing an increase in future income (Mutascu, 2021). Thus, the improvement of productivity indicators is due to more precise functioning and the fact that the technology is unlikely to get any errors.
Conclusion
In conclusion, the most significant limitations of artificial intelligence are rising unemployment and increasing laziness. Thus, it negatively affects society and its productivity in the workplace. On the other hand, others believe it is beneficial as it is faster and more efficient. Moreover, it contributes to improving performance since the innovative technology can work for an unlimited amount of time. Therefore, this argumentative essay concluded that artificial intelligence could be as used as an assistant to avoid people losing their jobs which could impact people’s lives.
References
Bordot, F. (2022). Artificial intelligence, robots and unemployment: Evidence from OECD countries. Journal of Innovation Economics Management, 37(1), 117-138. Web.
Korinek, A., & Stiglitz, J. E. (2018). Artificial intelligence and its implications for income distribution and unemployment. In The economics of artificial intelligence: An agenda (pp. 349-390). University of Chicago Press.
The appropriate handling of client complaints is unquestionably the most crucial aspect of offering exceptional customer service. Customers post unprompted reviews of one’s goods and services on online forums, social media, and pretty much anywhere else on the internet. And since customers feel most obliged to express their ideas when they are unhappy, spontaneous feedback is typically negative. These internet complaints are increasing daily and might be challenging to control. Artificial Intelligence (AI) and Machine Learning (ML) are the recent tools that can be used by government agencies to enhance the process of managing complaints.
Discussion
Giving machines access to data sources and allowing them to learn the knowledge without being explicitly taught is referred to as machine learning. Large amounts of adequately structured data are produced by customer service, for example, when consumers ask queries and support teams respond (Gacanin & Wagner, 2019; Marinchak et al., 2018). The solution is to classify customer complaints using machine learning. Wherever they may appear, client comments can be tracked by AI-powered text analysis tools to identify those that are complaints, route them automatically to the right team, and analyze them for immediate valuable insights (Roldos, 2020).
The industry tested tens of thousands of messages and offers in the fall of 2020 while collaborating with the AI start-up OfferFit, changing the creative content, channel, and delivery times. It redesigned its organization to focus on client acquisition, service, and renewal and started utilizing AI to schedule service calls more effectively, support call center agents’ cross-sell recommendations, and reach out to customers about upgrading their wireless systems (Edelman & Abraham, 2022). Brinks increased A/B testing from two or three tests per day to almost 50,000 tests in less than two years (Edelman & Abraham, 2022).
However, not all cases of the use of ML and AI technologies in managing customer complaints were successful. Giving machines access to data sources and allowing them to learn the knowledge without being explicitly taught is referred to as machine learning. Large amounts of adequately structured data are produced by customer service, for example, when consumers ask queries and support teams respond. According to F33 (2021), the harsh truth is that, despite the fact that artificial intelligence (AI) and machine learning (ML) are currently very trendy terms and that almost every tech company’s product and solution is AI-enabled, the majority of the customer’s entities have largely failed to implement ML within their own organizations. It is common for an enterprise to stagnate when there are too many ideas or prospects for one to evaluate (“How can AI,” n.d.). This could be due to a lack of interest in committing to one idea because it is likely that a better machine-learning project will emerge.
There are both benefits and challenges to the use of AI and ML in the customer complaint resolution process. A company’s capacity to deliver a customer experience that competes with the competition depends on its ability to provide quicker solutions, 24/7 support, and predictive learning (Vaught, n.d.). Today’s high expectations for customer service make it impossible for a company to ignore AI-powered support systems. AI integration isn’t always simple, though. An enterprise’s support staff will need to adjust in a variety of ways.
Institutions may maximize the potential of their extensive multilingual databases by using AI. They can also reach international markets more quickly. For example, language technologies like NMT make translation faster and less expensive. When it comes to translating vast volumes of text and detecting languages, no human translator can compete with a machine. AI allows greater scalability and scope, whether for gisting purposes or content intended for post-editing by human translators. A deep learning technique called NMT enables MT engines to train on their own. It employs a synthetic neural network, which is akin to how one’s brain functions.
Conclusion
In summary, the most critical component of providing excellent customer service is the proper management of customer complaints. Customer complaints can be categorized with the aid of machine learning. The ability of a company to provide a customer experience depends on that business’s power to offer quicker answers, round-the-clock service, and predictive knowledge. An entity cannot disregard AI-powered support systems, given the high standards for customer service that exist today.
Gacanin, H. & Wagner, M. (2019). Artificial intelligence paradigm for customer experience management in next-generation networks: Challenges and perspectives. IEEE Network, 33(2), 188-194.
Artificial intelligence has been the forefront of science and technology for decades. The term artificial intelligence is often misinterpreted as the recreation of human intelligence in machine form. However, it would be apposite to define it as an ability of an artificial system to gather, interpret, and apply data for the achievement of specific goals.1 A distinct feature of artificial technology is the ability to execute given tasks without human oversight and improve performance, with algorithms allowing it to learn from its mistakes and input data.2 Therefore, programs and techniques based on artificial intelligence are uniquely positioned to accomplish complex functions in a variety of fields.
Discussion
The idea of machines possessing and operating with human-like intelligence can be traced to the first half of the 20th century. In 1942, Isaac Asimov published a short story about robots possessing artificial intelligence and outlined the fundamental laws of robotics.3 Many scientists were inspired by Asimov’s story, commencing work on intelligence techniques. In 1956, the Rockefeller Foundation founded a workshop on artificial intelligence hosted by Marvin Minsky and John McCarthy, who are considered the fathers of this branch of science. 1956 is considered the birth year of artificial intelligence and the year the term was coined. Early artificial intelligence projects included the 1964 natural language processing program ELIZA and the 1959 General Problem Solver program aimed at the solution of universal problems.4 The 21st century produced more intricate artificial intelligence programs and techniques that are utilized in lethal autonomous weapon systems, intelligence, surveillance, and reconnaissance, as well as in medicine, logistics, education, and cyberspace.5 The latest leaps in artificial intelligence are embodied by quantum computing that allows for faster data gathering and interpretation.6 Quantum algorithms present a riveting field of research, particularly in intelligence collection and analysis.
Today, defense and intelligence agencies have unprecedented access to artificial intelligence programs and techniques. Although the artificial intelligence used or developed by the U.S. military and by the military-industrial complex is considered to be in its infancy, it can be highly beneficial.7 Artificial intelligence allows for faster data interpretation, translating into an increased speed of decision-making processes in the military and reaching more objective solutions by mitigating human error.8 Moreover, the use of artificial intelligence decreases human labor and costs.
Nevertheless, many experts argue that applying artificial intelligence, including techniques based on quantum computing, presents substantial security challenges. Specifically, the speed of artificial intelligence systems creates incentives for opponent states to resort to preemptive actions, leading to escalation of conflict.9 Moreover, there is an inherent risk of loss of human control over vital decisions if the system disregards data it marked to be inconsequential. For example, failure to specify certain conditions as dangerous can lead to the inability to plan a safe path in autonomous submersibles. Quantum computing, in particular, raises questions pertaining to security and efficiency, with a high potential for algorithm-related data. It should be noted that most resolutions utilized in artificial intelligence are based on heuristic algorithms that may not present suitable solutions.10 Thus, the supposed benefits of a more efficient decision-making process remain uncertain.
Quantum-based artificial intelligence and its use in the military, including intelligence gathering and interpretation, present an interesting field of research. The potential for mistakes in decisions and solutions proposed by artificial intelligence systems can lead to potentially devastating outcomes that can affect numerous people. Therefore, this paper aims to answer the following research question: how do quantum computing algorithms impact artificial intelligence in intelligence collection and interpretation? What is the potential for error and miscalculation in quantum algorithms?
Purpose Statement
Quantum technology can be applied in a variety of fields within the military. It can be defined as technology built with the use of quantum-mechanical properties, including quantum entanglement, superposition, and tunneling utilized in separate quantum systems.11 Thus, quantum warfare is the use of quantum technologies and artificial intelligence in support of the national security, at strategic, tactical, and operational levels, through the employment of highly advanced and efficient gathering and analysis of intelligence. This paper addresses the use of artificial intelligence systems based on quantum technologies in the military, specifically in intelligence data interpretation. Furthermore, its impact on U.S. national security will be assessed, with the paper considering the effect of data interpretation miscalculations on the nation’s ability to defend itself.
The comparison of artificial intelligence systems based on different technologies, including quantum technology, will help elucidate how data is collected, excluded, and evaluated by different systems. Unlike other technology, quantum tech utilizes quantum bits that hold more information than binary digits, thus, processing any data set at an increased speed.12 This will yield an understanding of how quantum-based artificial intelligence techniques operate and how they can benefit the U.S. intelligence agencies. Therefore, this analysis will help assess whether the investment in quantum technologies by the military, in particular, intelligence agencies, is justified.
Conclusion
Furthermore, assessment of the potential for errors and miscalculations of systems based on quantum technologies will allow evaluating their safety and efficiency. In discussing this question, both technological and ethical aspects of artificial intelligence implementation are to be considered. Such ethical principles as justified and overridable uses of artificial intelligence and human moral responsibility require exploration.13 The emphasis will be made on the possibility of miscalculations in implementing different algorithms. Nevertheless, the ethical issues arising from such errors should not be divorced from the conversation, as mistakes made by the military have the potential to impact people on the national level. Therefore, the study has two primary purposes:
To examine the efficiency of artificial intelligence systems based on quantum technologies compared with those not built on quantum technologies.
To consider the probability of errors in data collection and interpretation and their effect on U.S. national security.
Haenlein, Michael, and Andreas Kaplan. “A Brief History of Artificial Intelligence: On the Past, Present, and Future of Artificial Intelligence.” California Management Review 61, no. 4 (2019), 5–14.
Taddeo, Mariarosaria, David McNeish, Alexander Blanchard, and Elizabeth Edgar. “Ethical Principles for Artificial Intelligence in National Defence.” Philosophy & Technology 34, no. 4 (2021), 1707-1729.
Footnotes
1 Michael Haenlein and Andreas Kaplan, “A Brief History of Artificial Intelligence: On the Past, Present, and Future of Artificial Intelligence,” California Management Review 61, no. 4 (2019): 1.
Technological advancements have revolutionized social and economic growth due to automation and consequently increased output. Artificial intelligence (AI) is one of the greatest technological advancements that allow human intelligence simulation processes by machines (Larrañaga & Moral, 2011). Finance, healthcare, marketing, and fields can benefit the most from AI (Neapolitan & Jiang, 2018). Integration of AI in the fields would make life on earth easy for human beings through service improvements. Although AI can be significant for social and economic development, technology should be less involved in education.
Finance, healthcare, and marketing are crucial for routine human activities and survival. Integrating AI into financial systems can help avoid common human errors (Rico-Contreras et al., 2017). Moreover, AI, through predictive features, can help solve complex financial projections. Human well-being is significant for social growth. Therefore, using AI for medical research and advanced treatment will promote quality healthcare (Kautz & Singla, 2016). Meanwhile, marketing allows business organizations to reach a broad consumer base, increasing their profitability. Automating and digitizing marketing activities can be beneficial to corporations. Automation and digitization of healthcare services, marketing, and financial systems through AI are beneficial.
While AI is significant for social and economic growth, its application in education should be limited. The technology allows students who take technical courses such as engineering to advance their projects. Additionally, the technology helps institutions project students’ performance and organize teaching plans (Holmes et al., 2021). However, the use of AI in helping students improve their classes works is detrimental. For instance, systems that automatically solve class assignments encourage laziness among students. Moreover, AI systems that grammar check and rephrase students’ research work discourage creativity. Although AI is beneficial to educational institutions, it can lead to laziness and a lack of creativity among students.
References
Holmes, W., Porayska-Pomsta, K., Holstein, K., Sutherland, E., Baker, T., Shum, S. B., Santos, O. C., Rodrigo, M. T., Cukurova, M., Bittencourt, I. I., & Koedinger, K. R. (2021). Ethics of AI in education: Towards a community-wide framework. International Journal of Artificial Intelligence in Education. Web.
Many algorithms are used in AI, and they all have strengths and weaknesses. One of the most common is the support vector machine (SVM), which is used to classify data into one of many categories.
Discussion
SVM’s strength is that it can work with large datasets, but it has a high learning rate and requires large amounts of training data. Another standard algorithm is neural networks, similar to SVM, in that they learn from data like humans (Chang et al.,2018). Neural networks can be trained to recognize patterns in data, but they are also sensitive to noise and require more training time than SVM. Another type is a random forest classifier, which uses multiple decision trees to make predictions about new examples that have yet to be seen or have not been seen often enough for classification algorithms like SVM to work well on them.
One of the strengths of using one algorithm over another is that it can be more easily adapted to fit the needs of different problems. For example, if one is trying to find a solution to an optimization problem and has an algorithm that performs well in that environment, it is best to stick with it. Nevertheless, if one is working on a problem where many factors affect results, something else is recommended.
The strength of choosing one algorithm over another is that they tend to produce similar results for similar problems. This means that if one’s computer has access to one sort of algorithm and has already run it many times before without finding an optimal solution, it should be able to find one quickly using whatever variation given. However, The most significant weakness of this approach is that it can take quite a bit longer than other methods, especially when one is trying to solve problems that require human input or creativity.
Conclusion
There are a lot of different methods to choose from when it comes to AI Algorithms; some include reinforcement learning and support vector machine methods. I use the reinforcement learning algorithm since this method can teach me how to solve problems and make decisions without being told how. A reinforcement learning algorithm can be used in many different ways, such as when it comes to machine learning or robotics. The main reason I would use this method is that it allows for flexibility, which means it can be used in any situation or for any purpose.
Artificial intelligence’s methods for discovering solutions to problems can be implemented with and without understanding the domain, depending on the circumstance. AI decision-making has been studied from various functional and industrial viewpoints in academic and practical literature. As AI-based services continue to advance, more personal and important decisions are being left to the technology. However, the two fundamental principles of propositional and first-order logic are the foundation for all AI-based technologies.
The first core element is propositional logic, which uses Boolean reasoning to transform real-world data into a machine-readable structure. Such reasoning applies to knowledge-based expert systems or AI-based systems that make decisions or judgments similar to those of experts. Completeness, consistency, and tractability may be required to very effectively and efficiently formulate the domain knowledge as theory (Neapolitan & Jiang, 2018). Complex prepositions link one, two, or more sentences together. In propositional logic, the syntax used to indicate the joining of two or more sentences is created using symbols. An appropriate structure for presenting information is known as syntax.
Artificial intelligence’s propositional logic analyzes sentences as variables, and in the event of complicated sentences, the first phase is to deconstruct the sentence into its component variables. The stages that carry out the intended task are hard-coded in the script when a system is designed procedurally (Kakas et al., 2017). The knowledge and reasoning are divided into the declarative method. This strategy helps develop wumpus algorithms that reason with knowledge bases and when using forward and backward chaining to reason with rules. Hence, propositional logic in artificial intelligence is essential to realize the promise of machine learning and decision-making fully.
Another method of information processing in artificial intelligence is first-order logic. To concisely convey the natural language statements is an extension of propositional logic. A machine that uses a comprehensive knowledge base, such as First Order Logic, might be able to reason about a wide range of global issues. It could plan appropriately, reason from first concepts, argue about its objectives, and explain how its activities in the world interact with one another.
These computers can be seen performing these tasks in research centres and laboratories. In order to effectively account for the ambiguity or imprecision of the natural world, Garrido (2010) claims that the idea of sets, relations, and other ideas must be transformed by including logical concepts and procedures. Additionally, the potential for the abstraction of propositional logic is constrained because it does not permit the conduct of reasoning over variables and functions with generic and dynamic content. It implies that the early logical computing systems were likewise incapable of resolving issues whose solutions are included within the vector spaces whose subspaces the propositional space belongs to. First-order logic, a formal logical system that incorporates variables and enables abstraction, as a result, helped to solve this issue.
As a result, decision-making results in behaviors depending on the information and comprehension that the agent has been given. A portion of the risk associated with a decision is transmitted to the inputs of the decision-making agent if it directly affects the environment. In the area of artificial intelligence, the connection has received extensive research. Studies already conducted generally focus on constructing an AI’s reasoning frameworks using classical logic, or at least parts of it. However, first-order and propositional logic provides the fundamental knowledge for the continued development of AI systems.
References
Garrido, A. (2010). Logical foundations of artificial intelligence. Brain: Broad Research in Artificial Intelligence and Neuroscience, 1(2), 149-152.
Robotics and AI are widely used today in many companies around the world. Every day, technologies are becoming more and more sophisticated, making it possible to simplify tasks for personnel and increase processes’ efficiency. However, there is a downside to technological perfection – many people fear that AI and robots will supplant white-collar workers, as happened with the decline in blue-collar jobs during the Industrial Revolution. The question of AI’s influence on wider society is often discussed in terms of potential benefits or harms of the broader adoption of technologies. This essay hopes to reach an agreement on a presented question by examining people and technology’s interaction through the lens of organizational theories. In particular, it is proposed to analyze the possible opportunities of introducing technologies for workers and organizations, applying various approaches. Based on the analysis results, it will be possible to conclude how robotics and AI impact society.
Critically demonstrate critical awareness of knowledge issues related to the rise of Robots, and Artificial Intelligence (AI) on employees, organization and wider society.
Scientific Perspective on Robotics and AI
Scientists have different opinions on the positive and negative consequences of the widespread adoption of robotics and AI, but they adhere to one general direction in conducting research. In particular, Li et al. (2019) note the significant impact of employee awareness of AI on turnover intentions. Scientists emphasize that such intentions can be reduced by providing organizational support and creating a competitive psychological climate. Then, Wirtz (2018) examined the interactions of human employees and robotics in the service industry. The scientist concluded that AI services would likely be perceived by customers as commodities, mentioning the example of ATMs. According to the scientist, service robots are also unlikely to become the companies’ primary competitive advantage.
This conclusion suggests that the human-focused principle that considers employees as the main competitive advantage will be preserved. Simultaneously, AI data processing and outstanding AI training practices can become another factor in the company’s success. Interestingly, Wirtz (2018) acknowledges that “robots will master cognitive and analytical tasks of unprecedented complexity and will be able to mimic surface acting-type emotions” (p. 7). Therefore, the scientist believes that “cognitive and analytical tasks with low emotional or social complexity will increasingly be performed by service robots” (Wirtz, 2018, p. 7). Besides, services, mainly emotional or social, like psychologists’, nurses, or social workers, will be mostly delivered by humans. Otherwise, cognitively complex tasks and those demanding emotional intelligence will be performed by humans, with the support of robotics and AI. Therefore, an organization will need to learn how to balance AI- and human-performed jobs by implementing ambidextrous management approaches and practices.
The need to establish interaction will inevitably arise, as robots will begin to implement tasks previously performed by humans. For example, the scientist notes that in one Chinese bank, it was decided to cut 1,000 jobs, since the work performed by low-skilled call-center employees began to be completed by chatbots (Wirtz, 2018). The scientist predicts a sharp decrease in the need for low-skilled human labor in the service sector and increased requirements for remaining employees’ who will deal with exceptional cases. The scholar supposes that HR departments will develop personnel training practices, including work in robot-human teams, and ambidextrous management and organizational practices.
Further, Lazanyi (2018), after examining adolescent attitudes towards AI, concludes that humans did not have enough time to adapt and embrace AI. The scholar also notes that trust is a significant factor in increasing readiness for change, and acknowledges that people are not ready for robotic co-workers. Scientists Galloway & Swiatek (2018) drew attention to the fact that today AI and robotics are more often viewed in the context of their potential in task automation. However, scholars believe AI has broader applications, particularly in public relations and other industries.
Then, Erdélyi & Goldsmith (2018) took a broader perspective on the implementation of AI and robotics in many areas of modern life and proposed creating an international AI regulatory agency to regulate AI technologies and help develop national and international AI policies across the world. The need to make such an agency is dictated by new technology challenges, such as threats to personal data security, which leads to financial or political risks, or the proliferation of new AI-controlled weapons.
Further, Yawalkar (2019) describes the role of artificial intelligence in HR management. According to the scientist, AI helps to carry out many HR tasks, mainly when sorting of job-applications, analyzing speech patterns during the job interview, through digital software interviews, scheduling interviews, and work meetings. Besides, AI can reduce discrimination and favoritism and increase transparency in hiring. It also enhances employee-friendly practices such as the organization’s training, increasing the efficiency at the workplace. The scientist believes that if the HR department introduces AI into their everyday routines, this will increase the level of trust in AI among other employees. Then, Brougham & Haar (2018) note that futurists predict that by 2025, a third of existing jobs could be replaced with smart technology, artificial intelligence, robotics, and algorithms. According to the study, with greater AI awareness, employees showed lower commitment and satisfaction (Brougham & Haar, 2018). Therefore, this study speaks of the importance of employee trust in AI and organization.
Siau & Wang (2018) discussed the importance of trust, presenting the sequence of steps required to build initial trust and develop continuous confidence in AI. In particular, the initially built trust requires going through the stages of AI representation, changing AI image and perception, getting reviews from other users, ensuring transparency and ‘explainability,’ and continuous access to AI. Further, to develop ongoing trust in AI, it is necessary to provide AI usability and reliability, humans’ collaboration and communication with AI, social activities involving AI, and privacy protection. After these steps, the company can proceed to job replacement and AI’s goals congruence with the humans’ ones.
Critically evaluate the pros and cons of using Robots and AI based on your knowledge, experience, and concepts/theories learned in class.
McKinsey 7s Model
The McKinsey 7s Model describes how companies act in a coordinated and synchronized manner based on the organizational structure. Wirtz (2019) mentions: “An additional area of interest is the effective management of people-robot teams and what type of soft and hard skills will be required when AI becomes an integral part of decision-making processes” (p. 8). The McKinsey 7s Model implies the existence of seven indicators that characterize organizational processes: structure, strategy, skills, staff, style, systems, and shared values.
All of these elements are interconnected and grouped around the central piece of shared values. ‘Systems’ include a company’s business and technical infrastructure that establishes work processes and a decision-making chain (What is the McKinsey 7S Model, n.d.). Therefore, this model gives technology an equal role in business decision-making. Simultaneously, ‘skills’ reflect the company’s capabilities and competencies that allow employees to achieve their goals. This formulation implies that technologies are the company’s capabilities or competencies that employees will use to increase efficiency. In general, the McKinsey 7s Model characterizes a company’s internal processes and is used when shared values are being tested, for example, during a merger or acquisition of an organization.
70:20:10 Model
This model can be applied when creating AI training opportunities for employees. According to the 70:20:10 Model, workplace learning accounts for 70% of experiential learning, 20% of social learning, and 10% of formal learning (Developing world-class employees, 2018). As noted by scientists, organizations need to implement AI, as this increases their competitiveness in the market. Simultaneously, conventional training may not be enough to implement AI and robotics, as humans are not yet quite ready to perceive robots as ‘co-workers.’ Therefore, combining the trust-building steps with the 70:20:10 Model will yield the best results.
Hofstede’s Cultural Dimensions Theory
This theory makes it possible to adapt AI implementation in different countries, drawing on the dimensions inherent in various cultures. Hofstede’s Cultural Dimensions Theory assumes high and low power distance, collectivism vs. individualism, uncertainty avoidance index, femininity vs. masculinity, short-term vs. long-term orientation, and restraint vs. indulgence indicators. For example, Chinese organizations embrace hierarchy, collectivism, are comfortable with uncertainty, have middle level of power vs. nurture importance, futuristic and long-term orientation, and normative repression regarding satisfaction of needs (What is Hofstede’s Cultural Dimensions Theory, n.d.). On the opposite, US organizations are more egalitarian, individualist, comfortable with uncertainty, have a middle level of power vs. nurture importance, traditional and short-term orientation, and more prone to indulge needs’ satisfaction.
Big Five Trait Theory and Emotional Intelligence Theory
Big Five Trait Theory can be applied in selecting personnel able to cope with more complex service tasks in organizations where AI is used. According to the theory, people can be characterized according to the prevalence of five qualities – extraversion, agreeableness, openness, conscientiousness, and neuroticism (Cherry, 2020). The most successful service workers will likely have high levels of extraversion, agreeableness, openness, conscientiousness, and low levels of neuroticism. However, this theory’s application will be more relevant to services that require superficial emotional involvement. The selection of personnel for traditionally social or complex emotional services will require a different approach.
Moreover, AI and robotics are not yet ready to show emotional intelligence, the quality inherent in good leaders. Emotional intelligence involves the ability to control one’s own and others’ emotions by recognizing, understanding, and choosing what an individual thinks or feels (Emotional Intelligence Theory, n.d.). Therefore, emotional intelligence is based on self-awareness, self-management, social awareness, and social skills. However, over time, robots can learn to copy the behavior patterns of emotionally intelligent leaders.
Critically review and provide clear recommendations based on your analysis.
Recommendations
AI implementation in organizations often starts with the HR department. In particular, AI simplifies HR managers’ work thanks to chatbots, speech pattern examinations, and interview scheduling. The HR department is responsible for creating a favorable atmosphere in the team and ensuring ambidexterity when working in parallel with humans and robots. Implementing this task can be helped by applying business models to develop an effective management strategy. McKinsey 7s Model, 70:20:10 Model, Hofstede’s Cultural Dimensions Theory, Big Five Trait Theory, and Emotional Intelligence Theory may be the most adequate for the task.
The McKinsey 7s Model defines the role of technology in an organizational structure. Technologies perform decision-making tasks being an element of the company’s ‘system’ and enrich employees’ job methods as a component of the company’s ‘skills.’ Next, the 70:20:10 Model provides an opportunity to deliver employee training with adequate attention to experiential, social and formal learning. Hofstede’s Cultural Dimensions Theory is useful when implementing AI and robotics in different branch offices and companies worldwide. Big Five Trait Theory can help hire employees in the service industries where the most significant reductions will occur after introducing AI and robotics. Finally, Emotional Intelligence Theory enables companies to develop better software for robots by integrating the behaviors of effective leaders in robots’ operations.
Conclusion
Thus, if organizations can successfully integrate AI and human input, it will benefit the wider society. In particular, employees will be freed up for more exciting work that only a human can handle, leaving the machines to perform monotonous and tedious tasks. At the same time, clients will receive services of higher quality and in a shorter time frame. The main drawback may be job cuts for low-skilled ‘white collars,’ leading to a widespread increase in the need for education, and possibly greater diversity in human professions.
The implications of introducing technology into workflows will depend primarily on how carefully companies integrate AI and robotics. In particular, much will depend on the cooperation of AI and robotics with human employees. If HR departments of companies correctly present innovations and ensure building trust in the AI, this cooperation will occur at a high level. As a result, employees and companies will receive the maximum benefit and competitive advantage from technology adoption. Companies’ progress and their successful work with customers will lead to significant economic growth, which will allow states to solve the problem of unemployment and raise the level of education of those who need it.
References
Brougham, D., & Haar, J. (2018). Smart technology, artificial intelligence, robotics, and algorithms (STARA): Employees’ perceptions of our future workplace. Journal of Management & Organization, 24(2), 239-257.
Erdélyi, O. J., & Goldsmith, J. (2018). Regulating artificial intelligence: Proposal for a global solution. In Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society (pp. 95-101).
Galloway, C., & Swiatek, L. (2018). Public relations and artificial intelligence: It’s not (just) about robots. Public relations review, 44(5), 734-740.
Lazanyi, K. (2018). Readiness for artificial intelligence. In 2018 IEEE 16th International Symposium on Intelligent Systems and Informatics (SISY) (pp. 235-238). IEEE.
Li, J. J., Bonn, M. A., & Ye, B. H. (2019). Hotel employee’s artificial intelligence and robotics awareness and its impact on turnover intention: The moderating roles of perceived organizational support and competitive psychological climate. Tourism Management, 73, 172-181.
Siau, K., & Wang, W. (2018). Building trust in artificial intelligence, machine learning, and robotics. Cutter Business Technology Journal, 31(2), 47-53.
Wirtz, J. (2019). Organizational ambidexterity: Cost-effective service excellence, service robots, and artificial intelligence. Organizational Dynamics, 10, 1-8.
Yawalkar, M. (2019). A study of artificial intelligence and its role in human resource management. International Journal of Research and Analytical Reviews, 20-24.
The idea of the economy’s digitalization has long captured the whole world. Nevertheless, many countries’ governments are delayed or slowly introduced it into citizens’ ordinary life. There are many fears associated with digitalization, for example, the sharp reduction in jobs. However, artificial intelligence’s (AI) introduction foreshadows an increase in the economic component, more accessible work for humans, and discoveries in entirely different areas achieved through humans’ and automatic robots’ cooperation.
People’s Anxieties about Digitalization
An excellent worry for an unenlightened person in this topic is the fear of losing their job because now it can be performed by a robot. Many do not understand the direction of scientific and technological progress and have a poor understanding of technologies and their capabilities to improve human life in different spectra. The biggest fear in the economy’s digitalization is the loss of jobs, but this will only be in the short term and only if there is the uncontrolled use of artificial intelligence (Yampolskiy, 2020, p. 3). The answer to this fear will only be the subsequent increase in the number of jobs after the global elimination of jobs, the condition for which will be the skills of handling autonomous robotic systems.
The second fear of society is the need to update the centuries-old model of education. Such changes will lead to the fact that people will have to gain new knowledge in various fields related to economics, programming, logistics, and linguistics. Moreover, some professions will be completely eradicated due to the automation of their work. Simultaneously, there are many vacancies in healthcare, education, transport, logistics, and training. For example, Haenlein and Kaplan (2019) claim that “human employees will be increasingly occupied with feeling tasks since thinking tasks will be taken over by AI systems” (p. 5). It is this sector that will be able to provide significant employment growth in the future. There will be much higher standards and academic requirements for the new professions of those who will operate AI devices.
People are also concerned about the direction of scientific and technological progress. Many think that science that studying AI’s main aim is to obtain a conscious object, which may cause the world to take over. However, this is a big misconception because people should distinguish between “human-substituting innovations,” in which people will only become servants, and “human-improving innovations,” which improve a person’s sensory and motor capabilities (Haenlein and Kaplan, 2019, p. 7). As an example, it can be cited the medicine and the field of prosthetics development.
Ethical Issues Related to Artificial Intelligence
Moreover, one of the most critical points that baffle the spread of digitalization is related to the ethics of introducing artificial intelligence into production instead of people. The ethics question is still open because it is complicated to understand how ethical artificial intelligence in people’s places with their subsequent dismissal is. For example, Dignum (2018) concludes that “AI reasoning should be able to take into account societal values, moral and ethical considerations” (p. 1). There is an increase of ethical questions when discussing society’s division into a rich and poor class and the middle class’s disappearance. The middle class’s disappearance is the automation of many work processes associated with the middle class’s work.
AI and Job Security
All professions, despite their possible future digitalization, can be divided into two categories. The first is those professions that are more susceptible to the rapid introduction of artificial intelligence into their process. The second category is those who, due to specific characteristics, are least susceptible to the inclusion of automated actions in their process. The main characteristics that we highlighted in the lesson that make professions most vulnerable to inclusion in their work are simplicity, repetitiveness, time to complete the task, efficiency, and accuracy. Another critical factor is that some jobs can be replaced with artificial intelligence to reduce labor costs. Resilient characteristics we consider are creativity, communication, and ethical decisions making.
We discussed jobs that we have already had and ranked them according to the degree to which we assumed they could be automated (from one – easy to four – near to impossible). Lawyers, Artists, and Filmmakers are professions that are almost impossible to automate, while Software Testers and Accountants are professions that are being digitalized easily. In the second part of the task, we talked about the work we were providing. We concluded that each of them should be the least susceptible to digitalization. For example, even though Software Architect efficient structures could be made with AI, it would require some manual intervention to ensure creativity and decision-making. Web App Developer, Data Scientist, and Project Manager are also included in our chosen professions.
For the last task, we chose the profession of Customer Care Agent. We answered all the questions and realized that AI could already handle many of these employees’ functions. For example, AI robotic voice can communicate with clients and customers regarding their queries based on relevant keywords and expressions. The stakeholders are the IT engineers who built the AI into the business system, its customers, and the business itself. Of course, this system will have both pros and cons. The advantages for the business are evident – much fewer labor costs, and the advantage is that the buyer can be connected to a robotic agent at any time, and he does not have to wait for his turn. However, the disadvantages can also be significant; for example, some customers may not like the service provided by artificial intelligence. As a result, they will decide to refuse to use this business’s services; in this situation, they will lose everything: both customers and business. We decided to leave both the AI system and the live staff, and it will also be convenient for businesses to use AI to perform small tasks that are not related to customers. Furthermore, we also decided to leave the live staff to communicate with customers over the phone. Given all the decisions taken, we believe these are the best ideas to satisfy all stakeholders.
Conclusion and Action Plan
To sum up, it is essential to note what an important role and meaning this activity carried. If initially our thoughts were aimed at universal automation, now it is clear how carefully we need to approach this process. After all, we considered all possible options in this work and agreed on exactly where to use AI. It is also important to note that an essential role in a person’s work is his creativity and ingenuity. This assignment shows how significant it is to develop logical thinking, creativity, and the ability to be an ethical specialist in the future. The ability to feel and use feelings rationally, turning them into sometimes excellent solutions to problems, can demonstrate a person’s dignity and being a good employee.
Reference List
Dignum, V. (2018). ‘Ethics in artificial intelligence: introduction to the special issue’, Ethics and Information Technology, 20(1), pp.1–3.
Haenlein, M. and Kaplan, A. (2019). ‘A brief history of artificial intelligence: on the past, present, and future of artificial intelligence’, California Management Review, 61(4), pp. 5–14.
Yampolskiy, R. (2020). ‘The uncontrollability of AI’. To be published [Preprint].
Transportation, the business of moving goods and people from one location to another, has undergone several studies, research, experiments, and modifications to get to where it is now. In the year 1787, the steamboat became one of the most significant milestones in the history of transportation. Previously, people had to rely on animal-drawn carts to go about. Following that, key achievements in the transportation business included the introduction of bicycles in the early nineteenth century, automobiles in the 1890s, railroads in the nineteenth century, and airplanes in the twentieth century (Lǎzǎroiu et al., 2020). The transportation industry has progressed to the point where vehicles can navigate and move without the need for human intervention. The industry has benefited from technological breakthroughs in its quest for innovation and evolution. Artificial intelligence (AI) is one such cutting-edge technology that has benefited the industry. Leveraging AI in transportation helps the industry improve passenger safety, reduce traffic congestion and accidents, reduce carbon emissions, and lower overall financial costs.
AI has long since moved beyond its theoretical presence in research labs to become prevalent in people’s daily lives. And, for the most part, technology has succeeded in its objectives. In a nutshell, AI is a technology that encompasses machines with human intelligence (Verganti, Vendraminelli, & Iansiti, 2020). Machines with AI skills may imitate people, automate manual jobs, and learn on the fly like humans do. With the introduction of automation, repetitive and time-consuming jobs fall under the purview of AI. Furthermore, according to Verganti et al. (2020), AI-powered systems exhibit human intelligence and learn over time, implying that these machines will eventually be able to perform critical-thinking tasks and make decisions on their own. Businesses in the transport industry are making considerable investments to boost revenue production and remain ahead of their competition, recognizing the exceptional potential of Artificial intelligence.
Self-Driving Vehicles
Autonomous cars are one of the most innovative uses of AI innovation. These vehicles, which were previously only in science fiction, are now a practical reality. Although some people were suspicious of this technology during its early phases, autonomous cars have now entered the transportation industry. In Tokyo, self-driving taxis have already begun to operate (Yaqoob et al., 2019). However, for safety reasons, the driver now sits in the car to take control of the cab in the event of an emergency. According to the manufacturers of this driverless taxi, the technology will result in lower taxi service costs, thus increasing public transit options in rural places.
Similarly, the logistics industry in the United States is adopting autonomous trucks in order to gain several benefits. According to the McKinsey Global Institute, vehicles move 65 percent of products internationally (Lǎzǎroiu et al., 2019). With the introduction of self-driving trucks, maintenance and administration costs will be reduced by around 45 percent (Lǎzǎroiu et al., 2019). For the time being, most corporations are still undertaking pilot programs to make self-driving vehicles faultless and safe for passengers. As this technology advances, self-driving vehicles will acquire widespread acceptance and become commonplace in the consumer market.
Traffic Management
Another issue that individuals confront on a daily basis is traffic congestion. AI is now poised to fix this problem as well. Sensors and cameras placed on the road capture a massive quantity of traffic data. This data is then transferred to the cloud, where it will be analyzed and traffic patterns revealed using big data analytics and an Artificial Intelligence-powered system (Javaid et al., 2018). Data processing can provide valuable insights, such as traffic projections. Important information such as traffic forecasts, accidents, and road closures can be delivered to commuters. Furthermore, users can be alerted of the shortest path to their destination, allowing them to travel without having to deal with traffic (Abduljabbar et al., 2019). AI may thus be used to minimize undesired traffic, enhance road safety, and decrease wait time.
Delay Predictions
Flight delays are another major issue confronting air travel today. According to a study undertaken by experts at the University of California, Berkeley, the estimated cost of aircraft delays in the United States is 39 billion dollars (Sun et al., 2020). Flight delays, in addition to financial loss, have a detrimental influence on passengers’ travel experiences. Negative flying experiences can diminish the value of a transportation firm, leading to greater client attrition. To address these difficulties, AI comes to the aid of the aviation sector.
Using data lake technologies and computer vision, the sector can provide great service to customers by reducing wait times and improving their travel experience. Because everything from adverse weather to technological malfunction can cause aircraft delays, it is critical to provide flight data to passengers ahead of time to avoid excessive wait periods. According to Gui et al. (2019), continuous monitoring of airplanes may be carried out with the use of computer vision systems, reducing accidental downtime. Furthermore, Artificial Intelligence and machine learning components will analyze real-time flight data, historical records, and meteorological data. On-the-fly computation will aid in the discovery of hidden patterns, which will provide the aviation industry with important insights into other potential causes of aircraft delays and cancellations (Gui et al., 2019). This information may be provided to passengers, who can then organize their itinerary appropriately.
Drone Taxis
A drone taxi is one of the most intriguing and creative AI uses in transportation. According to Jat & Singh (2020), pilotless helicopters provide a novel approach to reducing carbon emissions, eliminating traffic congestion, and reducing the need for costly infrastructure investment plans Furthermore, Gui et al. (2019) state that drone taxis will enable passengers to get to their location considerably faster, reducing commuting time. Additionally, growing populations have put city planners under intense pressure to ensure wise urban planning and infrastructure construction while conserving scarce resources (Jat & Singh, 2020). Drone taxis may be the true solution to all of the issues that these municipal officials are attempting to address. The recent demonstration of an autonomous aerial vehicle in China, in which seventeen passengers enjoyed smart air mobility for the first time, is an excellent predictor of such future uses.
Shipping, Navigation, And Ports
As many firms still employ vessels to convey products, the last several years have been critical for the growth of the maritime logistics industry. Waterway transportation necessitates the analysis of a large amount of data in order to optimize shipping routes for ships of various sizes and carrying differing sorts of products (Alop, 2019). Artificial intelligence in transportation software enables organizations to collect particular data that aid in decision-making, enhancing shipment safety and vessel energy efficiency.
Many variables influence marine logistics and navigation, making route planning difficult without a comprehensive collection of data. It may be more efficient to have the ship under control on shore. For example, the MUNIN project investigated an autonomous warship that navigated aboard but was controlled from land (Abaei et al., 2021). This initiative advances the possibilities for leveraging automation in the transportation business to improve ship navigation and quality marine delivery.
Intelligent Train Automation
The railway sector was once the most inventive mode of transportation, and it still can be with the application of AI to improve management and operational processes. A collection of firms began developing driverless train prototypes for autonomous freight and passenger trains in 2018 (Singh et al., 2021). The integration of sensors, cameras, and radars with artificial intelligence transportation software enables the creation of the “train’s eyes” tool. By 2025, the business hopes to have a fully automated driverless train.
It is necessary for railway infrastructure to forecast potential failures. Operational intelligence enables the utilization of data from railway sensors for precise forecasting and repair suggestions. Furthermore, AI assists the railway sector in assessing long-term performance and identifying opportunities for development. Laing O’Rourke, for example, utilizes AI to cut logistics planning time down to 19 seconds. They can prepare for 23 days thanks to AI and transportation software solutions (Singh et al., 2021). In comparison, a person could only plan maintenance work for three hours and one day in advance.
AI has been one of the most amazing technological advancements in human history. Despite every fantastic creation so far, it is crucial to recognize that people have just scratched the surface of AI and that much more is left to be studied. The uses of artificial intelligence in transportation outlined above are only a taste of the possibilities and opportunities that the technology may provide.