What Every Business Executive Needs To Know About Artificial Intelligence and Machine Learning

Image by Gerd Altmann from Pixabay

Artificial Intelligence (AI) and machine learning (ML) have been hot topics the last few years, and for a good reason. With AI, we can do previously impossible or costly to do things. Artificial intelligence is also benefiting businesses by automating processes, saving time and money, and improving customer experiences. But what every business leader must know about AI and ML?

AI is already playing an important role in our lives, but it is just the beginning. It has the potential to change how we work, teach, live and interact with others. Automation will be the key to reducing human effort in the fourth industrial revolution. This is because automation can do tasks that humans cannot. For example, robots can take care of dangerous work that humans should not be doing. Also, algorithms and machines can work on routine manual tasks and free up time for humans. At the same time, there are legitimate worries about how AI will affect human jobs in the future. AI is not only capable of taking over routine manual tasks but also able to replace humans in knowledge-based professions such as doctors and lawyers.

As artificial intelligence and machine learning continue to grow in importance, the executives and leaders of companies will need to work hard and fast to keep up with the latest developments. There is a lot of potential for growth in this field, so it’s essential that any CEO or other leader be aware of what AI and ML offer. My goal here is to demystify AI and ML for business leaders. I will cover AI and ML definitions and differences, types of AI and ML techniques, economic impact, AI in organizations, business use cases, and rationale for building digital competencies in business.

Definition of Artificial Intelligence

Artificial Intelligence (AI) is the ability of a machine to mimic cognitive functions that are associated with human minds. This includes the ability to learn, reason, identify patterns and make decisions. There are four types of AI: automated intelligence; assisted intelligence; augmented intelligence; autonomous intelligence

Automated intelligence is a term that covers different ways for intelligence to be present in a task and refers to the automation of manual and cognitive tasks that are either routine or non-routine tasks. Automated intelligence does not require humans in the loop, and it is the automation of existing tasks.

Assisted intelligence is when machines help humans complete a task faster and better than they could alone. Assisted intelligence includes hard-wired systems that do not learn with interactions and require humans in the loop. Both automated intelligence and assisted intelligence focus on automating relatively simpler tasks with ML and analytics using structured data; together, they make the first wave of AI, the algorithmic wave.

Augmented intelligence refers to a machine providing additional data or information to a human to help that human make better decisions. Such systems are adaptive, i.e., they learn from their interactions with humans and the environment. Repeated tasks such as filling in forms and statistical analysis of unstructured data in semi-controlled environments such as robots in warehouses are examples of augmented intelligence. Augmented intelligence is the second wave of AI, the augmentation wave.

Lastly, autonomous intelligence is the ability of machines to perform tasks without humans. Such systems solve problems in dynamic real-world situations and focus on automation of physical labor, such as driverless cars. Autonomous intelligence is the third wave of AI, the autonomy wave.

Economic Value of Artificial Intelligence

The economic value of artificial intelligence is estimated to be worth $15.7 trillion by 2030, according to a report, Sizing the prize, from PwC. According to a 2019 MIT Sloan Management Review and BCG AI survey of over 2,500 executives, almost 9 in 10 executives believe AI offers an opportunity for their business. Still, only 18% of these executives have used this technology to generate revenue. McKinsey’s report “The state of AI in 2020” found that half of the respondents have adopted AI in at least one business function. Further, 22% of respondents reported at least 5% of EBIT attributable to AI. Lastly, revenue increases are more common in half of the business functions due to AI adoption, while costs decreases are less common. Therefore, AI is worth the investment to generate economic value.

Artificial Intelligence vs. Machine Learning

Artificial intelligence and machine learning are often confused with one another. The two concepts are not the same, but they play a significant role in today’s technology. Machine learning is a subset of artificial intelligence that focuses on predictive analytics by evaluating large data sets and extracting patterns from them. The difference between machine learning and artificial intelligence is that machine learning refers to programs that learn how to do tasks by themselves without any human input. In contrast, artificial intelligence is an attempt to teach computers how to do tasks like recognize speech or make decisions based on the information they are given.

Definition of Machine Learning and Types of Learning

Machine learning algorithms are techniques that allow AI systems to learn from data.

The machine-learning algorithms are divided into four different types: supervised, unsupervised, semi-supervised, and reinforcement learning.

The most common machine learning algorithm is called supervised machine learning, which works by feeding the computer data and then asking it to make predictions about future outcomes. Supervised learning is when we have a set of training data that has both input variables and the corresponding output variables. The algorithm trains on the training data to produce an inferred function that maps inputs to outputs.

Unsupervised learning is when we have input variables but no corresponding output variable. The algorithm tries to find hidden structures in the input data by using mathematical tools such as clustering and dimensionality reduction, which reduces the number of variables needed for describing an object or event.

Semi-supervised learning is a machine learning method that leverages both labeled and unlabeled data to improve the performance of supervised learning. Unlike traditional supervised learning, semi-supervised learning requires only a subset of the training data to be labeled. This can reduce the time and cost it takes to train an algorithm while still achieving near-optimal results. Semi-supervised learning algorithms are a subset of supervised learning algorithms. These models are typically used in the context of classification problems where the task is to predict which category an example belongs to based on its features.

Reinforcement learning is a technique in machine learning that is used to teach computers how to learn. It provides an environment where the computer can act, and its actions are reinforced or punished. The computer is given an environment with an objective function to maximize or minimize. The computer takes random steps in the environment. The reinforcement learning computer is given a reward when it achieves its objective function and a penalty when it fails. The computer tries to maximize the rewards and minimize the penalties. After many iterations, the computer learns how to maximize the reward.

Figure 1 illustrates major types of machine learning.

Figure 1

AI in Organizations

Artificial intelligence can be used in a variety of ways in a corporate setting. AI can be used to optimize internal operations and generate external customer value. Artificial intelligence allows companies to increase efficiency and reduce manual effort by automating processes such as document generation, data analysis, and customer service interactions. It also allows companies to generate external customer value by understanding customer needs and providing personalized services or offers. See below for common use cases of AI in business.

Figure 2 illustrates some common AI use cases.

Figure 2

To be at the center of competitive advantage, AI must somehow build organizational responsiveness. This can be done by making it easier for employees to do their own jobs with the help of AI. Moreover, AI can help give leaders more time to think about how to take their business in new directions. This requires breaking down top areas for change into delivery packages where AI can improve responsiveness. These small delivery packages could include new recommender systems, chatbots, and data-driven behavioral marketing tools. The goal here is to execute the vision and focus on outcomes that deliver benefits to customers as opposed to internal benefits. With AI techniques, we are transitioning from industrial era competitive advantage to digital era competitive advantage (Highsmith, Luu, & Robinson, 2020). From pre-digital to digital, the business goals have changed from ROI to customer experience (CX), and the technology goals have changed from cost/efficiency to speed/adaptability. While ROI and cost/efficiency remain important drivers, they are secondary to CX and speed/adaptability.

Businesses must adopt AI to stay relevant

Businesses often focus on their core competencies, leaving them unprepared for the future. The future is not always predictable, but businesses must prepare for the unknown. Focusing on your core competencies can leave a business unprepared for disruptions in technology that could change the way business will be done. Some businesses could be hesitant to implement AI technology due to a fear of the unknown and its impact on the current organizational structure and jobs. However, this fear does not diminish the fact that AI can be valuable for many businesses. For example, it has been shown that AI-powered chatbots can improve customer service and lead conversion rates. Executives can frame the discussions on creating new roles and improving efficiencies to free up time for workers to do other productive and creative tasks.

Time is of the essence if businesses want to stay relevant. Those businesses that had invested in digital technologies faired better during the COVID pandemic. The pandemic has acted as an accelerator for digital transformation, and AI technologies will continue to work for humans to improve productivity and speed of decision-making under uncertainty.


In the Fourth Industrial Age, agility and speed will be dominant forces for success. Businesses must adopt AI technologies that already exist rather than think about what AI could be. AI investments will require a portfolio and program management, and business executives must be much more responsive than they were in the past. While AI technologies may still rely heavily on legacy processes, however, the structure and process of earlier times may not be relevant to the digital organizations of the future. AI has the power to transform an entire organization and its processes. The path to successful digital transformation and developing a leading digital enterprise lies in being innovative, fast, value-centered, and adaptive.

Highsmith, J. A., Luu, L., & Robinson, D. (2020). EDGE : value-driven digital transformation (1st edition ed.). Boston: Addison-Wesley.
Sizing the prize, PwC Report
Will robots really steal our jobs? PwC Report
Deploying AI to Maximize Revenue. BCG article
The state of AI in 2020. McKinsey article

Image by Gerd Altmann from Pixabay