Artificial Intelligence: A Risk and Impacts Perspective

Today, artificial intelligence, or AI, is on the minds of most people regardless of generational issues. It is used as a cutting-edge technology or is simply fashionable, although it is not a new concept, with origins that date back to 1943. It has been evolving exponentially in recent years as additional technologies have emerged that together have allowed the simpler and easier use of artificial intelligence, enabling its extension to application in various fields, from personal to school, work, and business applications.

There are several considerations, ideas, or definitions of what artificial intelligence is, depending on the approach because it has many variants.For example, John McCarthy, one of the pioneers of artificial intelligence, considered it , “The science and engineering of making intelligent machines, especially intelligent computer programs.”Alan Turing, considered one of the fathers of modern computer science, said: “A computer can be called intelligent if it can fool a person into believing it is a human.” On the other hand, Jerry Kaplan, an expert in AI,  started giving it a more global vision based on data and the capacity it can have in terms of “making generalizations in a timely manner, based on limited data,” which is very much in line with what Marvin Minsky, co-founder of the Artificial Intelligence Laboratory of the Massachusetts Institute of Technology, considered “the science of making machines do things that would require intelligence if done by a human.” There are many more concepts or definitions in this regard, but a more general one considered by UNESCO (United Nations Educational, Scientific and Cultural Organization) is “machines capable of mimicking certain functionalities of human intelligence, including perception, learning, reasoning, problem-solving, language interaction, and even creative production.”

The various benefits of AI have already materialized, hence the global boom worldwide and the bet by many technology companies to be one step ahead, making large investments in research and development and even entering into major alliances. There is also more of a business approach for the purpose of generating higher income. In that regard, the use of artificial intelligence in companies has mainly sought to optimize or make operations more efficient, hand in hand with cost reductions. Some advantages in this regard, among many others, may be:

• Automation of processes, and repetitive or routine tasks through the use of robots, and simplifying client services through chatbots or interactive voice response processes.

• Reduction of errors and greater efficiency and precision in the execution of tasks or activities in production lines that required manual adjustments, detecting defects undetectable by the human eye using infrared sensors.

• Simplification or reduction of process times, mainly in terms of data management and analysis, which provides timely information,even in real time, for better decision-making, thus impacting corporate strategies. 

• Derived from the previous advantages, an additional advantage is created, which is an increase in operational productivity of machinery and equipment as well as labor productivity among human resources. 

• Improve controls with a greater capacity for validation and continuous monitoring, including reduction of risks, deviations or failures in a preventive and not corrective manner by detecting areas of opportunity or notifications of actions without dependence on the human factor.

• It is also possible to obtain benefits in terms of innovation and development by having a greater installed capacity in the operation since AI frees time for humans to analyze processes and generate ideas for improvement through creative thinking and business process reengineering, thus maintaining an upward spiral of optimization.

• Although all the advantages mentioned above are definitely of high value when considering the use of artificial intelligence, there are still certain gaps or gray areas in terms of the risks that its use may generate. Some risks may have very little impact depending on the use and environment, and the information that is being used or shared, including the intended purposes in each case. Other issues entail greater risk or impacts when using AI. These include:

• Cybersecurity: Regardless of artificial intelligence, the issue of the increased use of technology for remote processes as a result of the pandemic a few years ago strongly increased cyber risks. Having machines to ensure a company’s security culture is carried out properly can be of high risk, especially if the programming of any robot has any vulnerability.

• Data privacy: If there is no adequate control over the use of critical and confidential information, given that AI collects and analyzes large amounts of personal data, it can be exposed to theft, or even left unconsciously in systems that store the information and that could be exploited by other people.

• Social and economic impact: From a social point of view, one of the main risks is the possibility of AI systems being manipulated to create false news or to disseminate erroneous information, which can generate economic impacts due to financial behavior, and changes in the labor market, which can lead to the critical situation of generating massive layoffs.

• Biases in the analysis of information: Given the lack of human sensitivity, judgment, or even common sense or perception, AI systems can demonstrate unexpected behaviors or make incorrect decisions that lead to negative consequences, in such areas as discrimination, the application of justice, or medical health issues.

Myths and Realities

Identifying the advantages of the use of intelligence versus having a holistic vision and clear awareness of the impact that can be derived from the use of artificial intelligence has generated, as in other diverse trends, some myths and realities.

Perhaps the most important or well-known myth is that “artificial intelligence will allow machines to think, learn and act like humans.” This is definitely not something that can be considered to date, although there has been an exponential evolution in AI and what was originally considered has been changing over time. Yet, even with the evolution of technology, the way a microprocessor works is very different from that of our brain.

Another important myth or perception is that “artificial intelligence can be used in every process.” Artificial intelligence cannot yet be considered universal; it has been successful in several fields, but there is still resistance to use it in some processes where the impacts could be catastrophic. 

Regarding the realities of AI, here are two cases with the highest percentage of certainty:

• Artificial Intelligence is insensitive and can generate racism and/or discrimination. This has already happened.  As mentioned above regarding  risks or disadvantages, AI does not apply appropriate criteria, nor can it be empathetic or use intuition.

• Artificial Intelligence is going to end up doing all repetitive jobs. Although we know that artificial intelligence is not yet a universal application, its use is constantly increasing and the companies that already use it to a great extent have done so to optimize processes and achieve greater productivity, which is more easily achieved in repetitive processes through the use of robots. This will surely continue.

With the continuous evolution, use, and applicability of artificial intelligence, new myths and realities will continue to appear, as well as different thoughts, trends, or positions regarding AI. It is such a broad topic that generates a lot of information in all environments and in relation to different topics, including strategic or trendy issues. Some very interesting topics that are also highly applicable to AI include:  digital transformation, administration of data and its holistic management, and technology governance issues, including the use of artificial intelligence. Surely, each of these topics can generate tons of information, having been immersed in the use of artificial intelligence in all its various forms.

The issue of talent in artificial intelligence also needs to evolve.  In the very near future, people will need to master AI in order to cover critical positions in all organizations. To that end, educational institutions will need to implement adjustments to their curriculums, mainly in areas with a technological focus. Finally at the organizational level, human resources teams will need to work strongly with holistic strategies and with the support of senior management to attract, recruit, develop, and retain talent with these profiles..