EN 
FR 
AI is not a silver bullet. It is a tool that, when used correctly, can empower businesses to make smarter decisions, improve efficiency, and create new opportunities.
- Sundar Pichai, CEO of Alphabet Inc. (Google)
AI is a powerful tool for your organization to improve its capabilities and achieve its mission.
- Fal Diabaté
Sundar Pichai's characterization of artificial intelligence (AI) as a tool is indeed correct. Business leaders would do well to pause and internalize AI's essence as a tool. While AI possesses immense potential to enhance productivity significantly, it remains fundamentally a tool. But what exactly is this tool, and why is there such widespread excitement surrounding it?
Artificial intelligence is a branch of computer science that aims to make systems (comprising both physical machines and software) with the ability to think and act like humans. These systems are referred to as algorithmic models or simply models. The concept of constructing models to replicate human reasoning is not novel. Since the early 1930s, and perhaps even earlier, computing futurists have harbored visions of these human-like machines. The term "artificial intelligence" was coined by John McCarthy and his colleagues at a research workshop held at Dartmouth College in 1956. Similarly, the term "machine learning," which falls under the AI umbrella, was also coined in the 1960s. During this period, models consisted of statistical symbols, logical rules for pattern recognition, and decision trees.
The adoption of AI was impeded by two primary limitations: access to data and computing power. However, the advent of Moore's Law in computing, the democratization of the Internet in the mid-1990s, the rise of open-source software, and advancements in communication networking addressed these two main issues. To be more specific, the introduction of the iPhone in 2007 and the subsequent proliferation of social media ushered in an era where a myriad of data from various industries—ranging from manufacturing to healthcare—flooded the internet, providing an abundance of data to analyze. With this influx of data, modern-day algorithmic models, also known as large language models (LLMs), are poised to deliver instantaneous and meaningful outcomes.
These LLMs perform essentially two tasks: training and inference, to generate outcomes. Training involves the learning mechanism models derive from the fed data and the provision of desirable outputs or outcomes. There are two types of learning: supervised and unsupervised. In supervised learning, the model is provided with complex data alongside the expected output, and it is tasked with predicting the correct outcome. In unsupervised learning, the expected outcome is not provided to the model, which must identify and learn the patterns and relationships within the data to provide the expected output. Inference, on the other hand, refers to the process of drawing conclusions from pre-trained models. The more finely tuned the model, the more accurate the inferences it can make. In a simplistic analogy, AI training and inference can be likened to two systems with distinct functions: System A—training—pertains to reasoning, while System B—inference—involves deduction. These two AI systems collaborate to provide accurate outcomes, including pattern recognition, image classification, prediction of the next word in a text or speech (natural language processing), and improvements to decision trees, at a high level.
Economists refer to super cycles as lengthy periods of economic expansion, characterized by growing GDP for nations, higher demand for goods leading to higher prices, and lower levels of unemployment. While the early impacts of AI, particularly generative AI, on society show signs of job loss in many industries, it is hard to argue that AI will lead to a super cycle. Similar to, and perhaps more than, other super cycles such as electricity and the internet, AI will be used as a tool to affect all aspects of our lives. Individuals such as programmers, students, teachers, writers, and business leaders who are early adopters of generative AI have noticed a tremendous increase in their productivity. They can achieve more with less in a short period of time. If people can improve outcomes with these robust chatbots, one can only imagine what will come next.
AI will impact all businesses, and leaders should keep in mind that it is a tool—a powerful one that should be utilized to empower people and improve processes to serve customers better. Today's business press is inundated with discussions about AI, not as a tool, but as an end to a mean. The nuances and features of AI often get lost in translations, leading to confusion of non-technical people. Good outcomes are attributed to AI, while blame is placed on humans. A CTO of a major tech consulting company once said, "The playing field is poised to become a lot more competitive, and businesses that don’t deploy AI and data to help them innovate in everything they do will be at a disadvantage." While this quote is beautiful, its framing is wrong on a few fronts. For one, it suggests that AI deployment makes companies competitive and innovative. I would argue that people make organizations more competitive and innovative. AI, when designed well and properly implemented, can fuel organizations’ competitiveness and innovation. I don’t mean to pick on this CTO; the business press is full of quotes and comments of this sort. Most people take them for truth and parrot them to appear smart.
Leaders should not get sucked into today’s business buzzwords around AI, defining their organizations as “AI organizations” instead of “AI-powered organizations,” which are two different things. A more appropriate language should be mission-driven organizations that utilize AI in an end-to-end approach to serve customers. AI, like any tool, can be used for good or bad. The kitchen fork is a trivial example; it can be used to cut and eat a nice steak, it could also be used as a weapon to kill. Similarly, leaders should equally weigh the benefits of AI, such as increase human productivity, improve processes, and gain customer loyalty, as well as the drawbacks related to AI hallucinations (false outcomes that are hard for humans to detect), governance, deep fakes (fabricated news, images, videos, or texts), and other ethical issues.
I’ll close by stating this; Artificial Intelligence (AI) is a powerful tool that will impact all industries and affect our lives. Leaders should not let themselves be drawn into the deluge of AI buzzwords and confuse their organizations’ missions with AI. Instead, they should treat AI as a powerful tool that empowers employees with increased productivity and wellness, improves internal processes for smoother organizational operations, and serves customers and external partners with intentional simplicity. Former IBM CEO, IBM Ginni Rometty said it best: “The key to successful AI adoption in the enterprise lies in integrating human expertise with machine intelligence. It's not about replacing humans; it's about augmenting human capabilities.”
Until we meet again, keep exploring and exploiting AI.
Fal Diabaté
Managing Partner, Barra Advisory Group