Incorporating Human Input in Artificial Intelligence

Incorporating Human Input in Artificial Intelligence

Introduction

Artificial intelligence is increasingly being applied in different economic sectors, including healthcare, manufacturing, and customer service. Some of the primary roles of artificial intelligence in these sectors include task performance and decision-making. Whereas some of these decisions may still be performed by humans, artificial intelligence handles significant amounts of data within a shorter period. Given artificial intelligence’s relative advantage, some organizations incorporated AI technology as a replacement for human labor. Such organizations only experience short-lived gains due to a myriad of factors. Since AI is a relatively new concept, people have yet to develop complete trust in AI systems to solve some sensitive problems. Besides, AI fails to provide identical solutions even when working on similar data due to the non-deterministic nature of AI algorithms. With such weaknesses, AI may not provide optimal solutions solely, hence the need for human input. Human input will be required to provide creativity, social skills, and leadership, while AI algorithms provide speed and scalability in task performance and decision-making.

Background Information

As artificial intelligence gains traction in the global economy, the element of trust emerges. AI will be applied in decision-making in sensitive sectors like healthcare, which means life and death. Since AI algorithms are non-deterministic, they may offer varying diagnostic outcomes for the same input data, raising fears among patients and healthcare providers about the effectiveness of the technology’s recommendations (Asan et al., 2020). Other human factors like individual biases, education level, and previous interactions with AI may raise trust issues. Once doubt sets in, it becomes hard to trust AI’s decision-making abilities and may even hinder healthcare providers from fully applying AI in healthcare (Asan et al., 2020). Even if trust levels may be optimal, that does not necessarily mean AI will solve healthcare issues. Suppose healthcare providers trust that AI outperforms human capability. In that case, they are unlikely to subject AI to further scrutiny, which may result in a catastrophe if AI makes wrongful recommendations (Asan et al., 2020). That calls for the development of a balanced mechanism that matches human trust based on AI’s capabilities.

Another societal problem emanating from AI’s adoption revolves around job insecurity and the fear that the technology will render some jobs obsolete. Data from leading global corporations show that incorporating artificial intelligence in customer service produces better outcomes compared to when human agents are used. Also, further statistics show that about 54% of jobs in the European Union are likely to be replaced by artificial intelligence, whereas 9% of the total new jobs in the US will be AI-based (Abuselidze & Mamaladze, 2021). To explain the increased adoption of AI in doing traditionally human jobs, corporation leaders aver that AI systems can work independently of external factors. While humans may take sick or maternity leave, AI systems are independent of such disruptions, which makes them a preference for profit-seeking organizations (Abuselidze & Mamaladze, 2021). There is a widespread fear that AI will render some traditional human jobs obsolete, leading to growing fears about job losses.

Additionally, a growing body of research suggests that AI systems perform better than humans in different fields. In healthcare, AI may be applied in hospitals, research institutions, and clinical laboratories (Reddy, 2018). Unlike humans, AI approaches can now perform healthcare functions that were previously unrecognized or considered unavailable. By employing machines, AI can now use unstructured data such as videos, photos, and physician notes to make valuable healthcare decisions (Reddy, 2018). Such discoveries beg the question of whether AI systems perform better than clinicians in delivering healthcare. It also brings more confusion as to what the role of clinicians will be in the future. Since AI is gradually proving to offer better insights compared to human healthcare providers, there is a threat that the latter will be relegated.

Proposed Solutions

AI Training and Deep Learning

There is a need to incorporate human perspectives to solve some of the weaknesses associated with artificial intelligence. Deep learning (DL), one of the latest AI technologies, leverages media infrastructure, and human-machine interactions to train itself and benefit from human insights to improve performance (Nobrega & Rutkowski, 2022). The increased use of deep learning means that humans will no longer rely on AI insights to improve performance but the other way around. Deep learning is a human-aided form of AI that allows the technology to make maximum use of human insights to deliver better outcomes. Deep learning will go a long way to alleviate the deficiency of trust between humans and AI systems.

Training AI systems is key to bolstering their interactions with humans since such training improves their social intelligence, which AI lacks. For instance, when it comes to customer service, there is a need for extensive training to improve how AI interacts with customers (Nobrega & Rutkowski, 2022). Microsoft’s chatbot, Cortana, Apple’s Siri, and Amazon’s Alexa are some of the AI chatbots that have received extensive human training to enhance their performance. The training helped Cortana to incorporate a friendly, kind, and helpful tone while responding to customers, while Apple’s Siri became sassy, a characteristic preferred by Apple customers. Also, the Fintech sector introduced supervisory monitoring for its AI bots due to the sensitivity of the sector. AI training solves the problem of distrust between humans and AI while preserving human jobs.

Incorporating Human Explainers

AI explainers will also be useful as they interpret AI solutions for human decision-makers. AI explainers will be particularly useful in evidence-based industries such as law, medicine, and law enforcement. Since AI decisions are processed through an opaque mechanism, there is always the need to explain these decisions to laypersons (Juho, 2019). For instance, decisions made in law enforcement may be subjected to further scrutiny by the final decision-makers. An expert in the specific field will provide the basis for AI decisions, thus alleviating trust issues (Juho, 2019). Also, an explainer will address sensitive AI-based healthcare decisions with the objective of helping the patient and a physician understand the decision. Once explainers are incorporated into the AI infrastructure, the fear of job losses will be alleviated, and the patients and physicians can trust the outcomes.

Incorporating Sustainers into AI Infrastructure

Sustainers are concerned with artificial intelligence aspects of AI. They will monitor the functioning of AI systems for potential malfunctioning and cyberattacks. In this age where cyberattacks are on the rise, sensitive industries like the financial sector are at risk of potential attacks. There is a need to introduce sustainers who monitor and report any possible attacks to alleviate the problem (Juho, 2019). In Fintech, customers often dedicate huge amounts of investments and rely on AI bots to manage those investments. However, most investors are always anxious about possible mismanagement, which may lead to huge losses. To that end, sustainers will offer supervisory control and assure users that their investments are safe (Juho, 2019). Apart from securing AI infrastructure, the incorporation of sustainers in the AI infrastructure preserves traditional human jobs that AI bots would otherwise take up.

Statistical Data Interpretation

Abuselidze and Mamaladze (2021) present data on AI’s impact on employment. To show that AI is penetrating the employment sector, the authors aver that 27% of companies believe they can deliver better outcomes by relying on AI bots in customer service instead of human agents. Further, they indicate that 48% of experts believe digital agents and AI bots will take up white-collar and blue-collar jobs. 9% of new jobs in the US and 54% of current EU jobs will be taken up by AI going forward. The article is unbiased since the authors present an unbiased view of AI by presenting an evidence-based review of AI’s employment interruption. On reliability, the authors discuss the utilization of AI in different sectors, with particular emphasis on customer service. A notable strength of the article is that research materials are referenced. However, the authors fail to mention the areas that need further research.

On the other hand, Nobrega and Rutkowski (2022) present data that focuses on the impact of AI on employment in Europe. A study survey shows that 74% of Europeans believe that current and new jobs will disappear due to the introduction of AI. Besides, 72% of the respondents believe that AI will replace people from their jobs, while 44% of the respondents believe that AI systems may partially complete their current job. The article is unbiased since it presents study outcomes from a survey conducted on a random sample of respondents. Since the article is backed by external research, it is reliable. One key strength of the information included in this article is that it is based on primary data collected by a reputable body.

Finally, Juho (2019) presents data to show the impact of artificial intelligence on ethical standards. He avers that organizations with over 250 employees are subject to ethical guidelines. One of the requirements is that such an organization should employ a representative responsible for data protection. The information presented in the article is reliable and unbiased since it draws from reliable organizations. However, a key weakness noticeable in the article is a failure to reference articles from where the data is drawn from.

Ethical Outcomes

The adoption of AI in healthcare diagnosis represents the positive ethical outcomes of artificial intelligence. Currently, most healthcare systems across the globe are grappling with rising healthcare costs and limited access (Vinay Kumar Bhargava, 2006). To eliminate the problem, AI has been incorporated into healthcare through deep learning and human-machine interactions. One key healthcare issue that AI will solve is the diagnosis of cancer patients (Morley et al., 2020). With the rising cases of breast cancer, patients face long waiting periods before being attended by physicians for diagnosis. However, with the adoption of AI into healthcare, early diagnosis is possible, and this will go a long way to save the lives of many patients. AI is a self-learning resource that may be used on demand to improve healthcare outcomes globally.

On the negative side, adopting AI in the financial industry poses serious risks. Due to the increased penetration of AI-based financial technology (Fintech) firms that face limited regulation, traditional banks have had to respond by lowering their capital investment. Since traditional banks are subjected to entity-based regulation, the increased competition from the Fintech sector means that banks cannot compete on the ground (Varma et al., 2022). Some banks have been forced to lower the amount of money they lend as loans, having lost customers to the Fintech sector. Besides, the consumer loses since Fintech firms tend to ignore ethical requirements on non-discrimination. To that end, minority populations have been denied access to financial credit since AI-based systems ignore key societal issues.

Conclusion

After considering the impact of artificial intelligence on different sectors, it is clear that there is a need for AI-human collaboration to attain optimal outcomes AI. AI is a relatively new concept in most sectors, particularly healthcare. The novelty means there is a lower trust level in AI systems to provide feasible solutions. There is also the concern of potential job losses resulting from AI adoption. Unlike humans, AI is not subject to external factors that affect performance, such as sickness; hence, most global profit-making corporations would readily accommodate AI in their operations to replace human employees. However, AI is subject to performance risks due to the application of non-deterministic algorithms, which may bring errors. Research has proved that there is a need for human-AI collaboration, which essentially alleviates these problems. AI will provide speed and quantity scalability, while humans will offer creativity, leadership, and team spirit for optimum outcomes.

References

Abuselidze, G., & Mamaladze, L. (2021). The impact of artificial intelligence on employment before and during a pandemic: A comparative analysis. Journal of Physics: Conference Series, 1840(1), 012040. https://doi.org/10.1088/1742-6596/1840/1/012040

Asan, O., Bayrak, A. E., & Choudhury, A. (2020). Artificial Intelligence and Human Trust in Healthcare: Focus on Clinicians. Journal of Medical Internet Research, 22(6), e15154. https://doi.org/10.2196/15154

de Nobrega, K. M., & Rutkowski, A. F. (2022, January). The AI Family: The Information            Security Managers Best Frenemy?. In HICSS (pp. 1-10).

Juho, V. (2019). Ethics of AI technologies and organizational roles: Who is accountable for the   ethical conduct? In Conference on Technology Ethics.

Morley, J., Machado, C. C. V., Burr, C., Cowls, J., Joshi, I., Taddeo, M., & Floridi, L. (2020). The ethics of AI in health care: A mapping review. Social Science & Medicine, 260, 113172. https://doi.org/10.1016/j.socscimed.2020.113172

Reddy, S. (2018). Use of Artificial Intelligence in Healthcare Delivery. EHealth – Making Health Care Smarter. https://doi.org/10.5772/intechopen.74714

Varma, P., Nijjer, S., Sood, K., Grima, S., & Rupeika-Apoga, R. (2022). Thematic Analysis of Financial Technology (Fintech) Influence on the Banking Industry. Risks, 10(10), 186. https://doi.org/10.3390/risks10100186

Vinay Kumar Bhargava. (2006). Global issues for global citizens: an introduction to key development challenges. World Bank.

ORDER A PLAGIARISM-FREE PAPER HERE

We’ll write everything from scratch

---

Global Societal Problem, Argument and Solution
[WLOs: 1, 2, 3, 4, 5] [CLOs: 1, 2, 3, 4, 5]
Prepare: Prior to beginning work on this assignment, please review this Sample Final Paper GEN499. Download Sample Final Paper GEN499 for additional guidance on the expectations of this assignment.

The topic of your essay needs to be a global societal problem from the following list:

Climate change
Pollution
Religious conflict and violence
Rise of artificial intelligence
Lack of education
Unemployment and lack of economic opportunity
Government accountability and corruption
Food and water security
International drug trafficking
Poverty and income inequality
Reflect: Based on the topic that you have chosen, you will need to use critical thinking skills to thoroughly understand how this topic can be a global societal problem and determine some logical solutions to the problem.

Write: This Final Paper, an argumentative essay, will present research relating the critical thinker to the modern, globalized world. In this assignment, you need to address the items below in separate sections with new headings for each.

In your paper,

Identify the global societal problem within the introductory paragraph.
Conclude with a thesis statement that states your proposed solutions to the problem. (For guidance on how to construct a good introduction paragraph, please review the Introductions & Conclusions. Links to an external site. from the University of Arizona Global Campus Writing Center to an external site..)
Describe background information on how that problem developed or came into existence.
Show why this is a societal problem.
Provide perspectives from multiple disciplines or populations so that you fully represent what different parts of society have to say about this issue.
Construct an argument supporting your proposed solutions, considering multiple disciplines or populations so that your solution shows that multiple parts of society will benefit from this solution.
Provide evidence from multiple scholarly sources as evidence that your proposed solution is viable.
Interpret statistical data from at least three peer-reviewed scholarly sources within your argument.
Discuss the validity, reliability, and any biases.
Identify the strengths and weaknesses of these sources, pointing out limitations of current research and attempting to indicate areas for future research. (You may even use visual representations such as graphs or charts to explain statistics from sources.)
Evaluate the ethical outcomes that result from your solution.
Provide at least one positive ethical outcome as well as at least one negative ethical outcome that could result from your solution.
Explain at least two ethical issues related to each of those outcomes. (It is important to consider all of society.)
Develop a conclusion for the last paragraphs of the essay, starting with rephrasing your thesis statement and then presenting the major points of the topic and how they support your argument. (For guidance on how to write a good conclusion paragraph, please review the Introductions & Conclusions. Links to an external site. From the University of Arizona Global Campus Writing Center to an external site..)

The Global Societal Problem, Argument, and Solution Paper

Must be 1,750 to 2,250 words in length (approximately between seven and nine pages, not including title and references pages) and formatted according to APA style, as outlined in the University of Arizona Global Campus Writing Center’s APA StyleLinks to an external site. Resource.
Must include a separate title page with the following:
Title of paper
Student’s name
Course name and number
Instructor’s name
Date submitted
For further assistance with the formatting and the title page, refer to APA Formatting for Word 2013 to an external site.
Must utilize academic voice. See the Academic Voice links to an external site. Resource for additional guidance.
Must include an introduction and conclusion paragraph. Your introduction paragraph needs to end with a clear thesis statement that indicates the purpose of your paper.
For assistance in writing Introductions & ConclusionsLinks to an external site. as well as Writing a Thesis StatementLinks to an external site., refer to the University of Arizona Global Campus Writing Center resources.
Must use at least eight scholarly sources.
Source Document Requirements:
Multimedia sources (such as videos) may be used, but no more than two such sources may be used. If multimedia sources are used, they must be authored and distributed by credible sources, such as universities, law schools, medical schools, or professors, or found in the University of Arizona Global Campus Library.
Government sources may be used, but no more than two such sources may be used. Examples include whitehouse.gov, state.gov, usa.gov, cdc.gov, and so forth. These websites can be used to make a stronger point about your proposed solution within the argument.
Where documents are used for source materials, those must be peer-reviewed, scholarly journal articles, and academically published books. Popular media sources (e.g., newspapers, magazines, television and radio shows, etc.) must not be used. Materials from advocacy groups (e.g., Greenpeace, Human Rights Campaign, National Organization for Women, etc.) must not be used.
Sites such as ProCon.org and Wikipedia must not be used.
Religious texts must not be used.
The Scholarly, Peer Reviewed, and Other Credible SourcesLinks to an external site. The table offers additional guidance on appropriate source types. If you have questions about whether a specific source is appropriate for this assignment, contact your instructor. Your instructor has the final say about the appropriateness of a specific source for an assignment. The Integrating ResearchLinks to an external site. The tutorial will offer further assistance including supporting information and reasoning.
Must document in APA style any information used from sources, as outlined in the University of Arizona Global Campus Writing Center’s Citing Within Your PaperLinks to an external site..
Must have no more than 15% quoted material in the body of your essay based on the Turnitin report. References list will be excluded from the Turnitin originality score.
A separate references page must be included that is formatted according to APA style. See the Formatting Your References ListLinks to an external site. Resources are available at the University of Arizona Global Campus Writing Center for specifications.
Good Critical Thinking Tips:

Your paper should include academic sources that explain multiple sides of the issue.
Your interpretations of the evidence should be objective and state the conclusions and theses presented in the evidence clearly and fairly.
Your paper should place the various forms of evidence in relation to one another and demonstrate why one form or perspective is stronger than the other positions that one could take on the issue.
Your paper should point out the limitations of current evidence and attempt to indicate areas for future research.