As artificial intelligence (AI) becomes increasingly integrated into various industries and applications, professionals must develop a foundational understanding of this transformative technology. This two-part article will provide an overview of the key principles, capabilities, and ethical considerations underlying AI systems, equipping professionals with the knowledge necessary to navigate the evolving landscape of AI-driven innovation and its implications for the industry.
The CEO of a multinational company sat by the owner’s side of her car as she commuted to work. There are always a few calls to take. While on her way, the car suddenly swerves into a park full of kids. Unfortunately, a few kids were hurt as they ran for their safety. Apparently, the autonomous vehicle had to make a split-second decision not to hurt the cat that ran into the road. Sadly, it was a poor decision as it put human lives in danger—both the CEO’s and the kids playing in the playground. This is a classic ethical dilemma in the context of autonomous vehicles. It’s commonly referred to as the “trolley problem” and raises challenging questions about how self-driving cars should be programmed to handle life-or-death situations.
Artificial intelligence algorithms power autonomous cars. The AI driver has been trained on millions to billions of data parameters. However, will it ever make human-like decisions on the wheel? Before we delve deeper into AI, it’s important to give you some foundational knowledge. To get started with AI, you do not need many technical details; you just need enough information to help your case.
Read also: Artificial Intelligence (AI) –What impact will it have on education?
Artificial Intelligence, Machine Learning, and Deep Learning
Most of the time, these buzzwords are used in place of each other and have been mentioned in a tremendous number of articles and blogs. What do these words actually mean technically?
Artificial intelligence
The term “Artificial Intelligence” was first coined by John McCarthy in 1956 during the Dartmouth Conference, a groundbreaking event that is widely considered the birthplace of AI. McCarthy, along with other pioneering researchers like Marvin Minsky and Claude Shannon, proposed the study of how machines could use language, form abstractions, solve problems, and improve themselves – establishing AI as a distinct academic discipline and setting the foundation for future research. While the term was new, the underlying concepts and work that would lead to AI had begun much earlier, including Alan Turing’s seminal contributions in the 1940s and 50s, as well as the development of cybernetics and early neural network research in the same era. The Dartmouth Conference marked a pivotal moment in the history of AI, solidifying its emergence as a transformative field with the potential to revolutionize how we think about intelligence and problem-solving. These handful of pioneers from the nascent field of computer science started asking whether computers could be made to “think”—a question whose ramifications we’re still exploring today – the Artificial General Intelligence concept.
As the name artificial intelligence suggests, it’s a concept that aims to automate intelligent tasks carried out by a human. This does not necessarily need to learn from data, it can be simple programming to turn on the light switch when it is 6 pm and turn off when it is 9 am every day. AI in its nascent era was rule-based programming. But of course, rule-based programming has a lot of limitations. For instance, what if the weather is so gloomy and everywhere is still dark by 9 am, can the AI decide to leave the lights on?
Machine Learning Although machine learning started to flourish in the 1990s it has a long history. The usual way to make a computer do useful work is to have a human programmer write down rules—a computer program—to be followed to turn input data into appropriate answers, just like Lady Ada Lovelace (regarded as the mother of computer science) writing down step-by-step instructions for the Analytical Engine to perform in the early 1840s. Machine learning turns this around: the machine looks at the input data and the corresponding answers and figures out what the rules should be. Going back to the light switch analogy, rather than program the switch to be turned on by 6 pm and off by 9 am, A machine learning algorithm is fed with data on when the light is turned on and off daily, and from the data, it can decipher the rule or when it should turn on and off the light.
In the second part of this article, I will explain deep learning, the foundation of the mighty LLMs (Generative AI) like Chatgpt, Gemini, and Claude, which we now use to carry out the vast majority of our creative and repetitive tasks.
Jonathan is a driving force behind democratizing AI in Africa. As Founder and CEO of Zummit Africa and co-founder and CEO of Datarango. He is empowering thousands with in-demand AI skills. Jonathan brings technical expertise as an AI developer with over 4 years experience to his mission. Beyond education, he’s also shaping the ethical landscape of AI as a board member with the Nigerian branch of the International Law Association.
Join BusinessDay whatsapp Channel, to stay up to date
Open In Whatsapp
