Essay Question: Given the potential moral risks of AI, should we stop their development?

Introduction

Artificial Intelligence (AI) has been achieving many things in the past few years in the field of healthcare, technology and business development. With the help of AI development, there are countless benefits ahead for economic growth of countries, improvement in safety and human wellbeing as well as social development. However, its complexity and capability as well as the level of explanation, security and privacy of information as well as moral dilemmas can possess potential threats and risks to significant users as well as economies (Siau and Wang, 2018). Many researchers disagree with the notion of Machine ethics and to be worried about it since AI is more of a developmental stage concern right now and rushing towards stopping its development will be a blunder (Alotaibi, 2018) whereas on the other hand, ethics and morality concerns should not be taken lightly and it can be assumed that sooner than later, AI will have similar consciousness to that of humans (Liang et al., 2021). This essay discusses the potential moral risk factors of Artificial Intelligence or machine learning as well as a few benefits to argue whether there should be a stoppage on its development or not. 

Potential Risks and Impact of AI

There are various ethical issues which are related to AI implementation out of which some major issues are discussed in this essay which are of highest concern. However, not all issues have same significance and importance in terms of using AI. Taking example of robots being used in a manufacturing process is not similar to those being used in healthcare or even working at nuclear facilities in relation to human and environmental safety (Sparrow, 2007). However, it is necessary to take into consideration the cultural and moral aspects of the society as well as regulations and norms if related to implementation of autonomous intelligence systems. This kind of factors may let AI systems behave unequally towards a community or even a huge part of population which is one of the main concerns. 

Unemployment caused by Automation

Considering ongoing achievements in the field of AI and advanced mechanics, it appears there is just a question of time until, even complexed job positions requiring high insight could be thoroughly taken over by the machines (Ma, Zhang and Zhang, 2018). If machines become faster, stronger, reliable and less expensive than human experts in numerous spaces of work, this would likely be the reason that the labour market to be evacuated on a scale not seen since the Industrial Revolution (Alotaibi, 2018). Mechanical advancement will augment the pay hole even further and may prompt falling wages and rising joblessness in huge fragments of the population. The individuals who will profit the most from innovative progress are individuals and countries that see how to utilize new innovative freedoms and the relating surge of massive information.

Specifically, nations with very much prepared IT experts are normal to succeed even with innovative advancement. All things considered, a full comprehension of the ways in which different calculations contrast with human dynamic and working capacities, just as the disadvantages of each, will turn out to be progressively significant later on, along these lines requiring elevated expectations of training. Furthermore, there is endless concern that AI and related advances could create huge unemployment during the following decades. Many researchers have that new data advancements will put a considerable portion of work, across a wide scope of occupations, in danger soon (Bird et al., 2020; Raso et al.,2018).

Inequality Concerns

The effect of AI on pay imbalance might be negative because of the differential effect it has on various positions and various specialists. Researchers found from an investigation of six EU nations that more youthful laborers, men, and those with mid-level training are more prone to be unemployed via automation (Panch et al., 2019). Furthermore, two significant channels through which AI-robotization will demolish pay circulation are recognized (Wang and Siau, 2018): one is through the developing advancement rents from AI for example the advantages to AI may just gather to a modest number of organizations; and a subsequent channel is through AI changing the overall interest for work, and subsequently change relative wages. 

Be that as it may, as on account of the effect of AI on jobs, AI may likewise, maybe strangely, improve the circulation of pay. This clarifies that significant level thinking requires practically very small calculations while low-level sensor-engine abilities require huge computational assets (Gries and Naudé, 2018). In this way, it will be complex for new innovation to displace the tasks and jobs that employees in the lower-expertise level jobs perform, such as, cleaners, nursery workers, receptionists or cooks. 

Another positive effect of AI on pay distribution might be, that AI may lead in reduction of the pay gap between genders and races since the high-ability automation will be in place (Panch et al., 2019). In their model of automation there are both high-ability work and low-talented work with the likelihood that if just high-skilled automation happens that the result might be neither efficiency nor inequal treatment increases (Sparrow, 2017). The instinct behind their outcome is that automation of expert work would diminish high salaries too, raising the cost of capital (over short term, as more companies will replace laborers with machines) and therefore reduction of productivity benefits.

Information Security and Privacy

The more remarkable an innovation turns into; the more would it be able to be utilized for negative reasons. This applies not exclusively to robots delivered to replace human soldiers, or self-sufficient weapons, however to AI frameworks that can cause harm whenever utilized vindictively. Since these battles will not be battled on the landmark just, online protection will turn out to be significantly more vital (Sparrow, 2007). All things considered, we are dealing with a framework that is quicker and proactive than an average huma being by significant degrees. In spite of the fact that AI is extremely amazing, it is as yet a moderately new innovation with regards to network protection, and still has a few impediments (Alotaibi, 2018). 

AI adequately is not refined to replace humans, however, there is yet the potential for false positives and missed discovery. Computer based intelligence is best utilized as an instrument for human experts to utilize (Ma, Zhang and Zhang, 2018) as it lessens the measure of time and assets that should be spent assessing likely dangers – however not an apparatus to replace individuals yet. 

Digital dangers are continually evolving since people on the other side are very creative and have for all intents and purposes limitless assets; in certain spaces, cybercrime is an economy unto itself. As new dangers arise, security arrangements that utilize AI must always be trained up to date in order to keep an upper hand (Bird et al., 2020). 

Cybercriminals can obtain AI-driven security systems and test their malicious projects to break them. Subsequently, they can hypothetically also create AI proof malware systems as well. They additionally use AI to comprehend what these security frameworks are searching for, and afterward can either mask their assault so their assault gives off an impression of being kind. Security is the solitary field where AI frameworks retaliate (Bird et al., 2020).

The precision of AI systems is not entirely perfect yet since these systems are not advanced enough yet in identifying between a malicious or beneficial activity. To ensure an organization and its applications and information, most IT security protection arrangements – including AI-put together arrangements – fail with respect to the side of alert. That is, if all else fails, banner something as bizarre and possibly undermining (Liang et al., 2021). This makes alarms about odd exercises that should be researched by human investigators and end up being considerate. The option in contrast to being careful to limit the quantity of false positives chances missing genuine assaults.

Should we stop the Development?

There are many potential risk factors related to AI which need to be considered but does it mean that its development should be terminated? There are many advancements in several industries which has benefitted and further will in the future which no one could have imagined without machine learning. The benefits of AI outweigh the potential risk factors which have only been proposed by researchers as precautionary measures or possible outcomes (Ma, Zhang and Zhang, 2018). Based on the research and self perception regarding development of AI, the benefits are far more valuable especially in the field of healthcare and human development than others which should be considered first and the development should be continued. However, in terms of ethicality and morality of the matter, certain policies should be practiced by authorities as well as individuals involved in order to keep it safe environment which lets human beings be in control at all times.

Conclusion 

In conclusion, as discussed above that there are several potential risks and ethical conundrums and concerns related to the development of Artificial intelligence being implemented in our industries, markets, healthcare systems and even daily lives, and misconceptions and confusions are present in everyone’s mind. However, if humans have created AI, they are well aware and capable of being in control at the same time as well. Therefore, to conclude, there are possibilities of threats and potential moral risks related to AI systems but the benefits are far more valuable and extensive which should not be overlooked just based on ethicality of certain cases and the development should not stop but precautionary measures should also be considered constantly.