Artificial Intelligence (AI)
Artificial Intelligence (AI) is the field of computer science focused on building systems that can perform tasks that would normally require human intelligence — things like understanding language, recognizing images, making decisions, and solving problems.
What is Artificial Intelligence (AI)?
For most of computing history, software did exactly what it was programmed to do. Write a rule, get a result. The machine had no flexibility — if a situation wasn't covered by the code, it had no way to handle it. Artificial Intelligence changes that relationship. Instead of always spelling out every rule in advance, many modern AI systems learn from data, identify patterns, and make judgments on their own. The result is software that can adapt, generalize, and handle situations its creators never explicitly anticipated.
The term AI covers an enormous range of technologies. At one end you have narrow systems built to do a single thing exceptionally well — the program that can beat the world's best players at a game like chess or Go, the weather service that forecasts tomorrow's rain, the thermostat that learns your daily routine and warms the house before you wake. At the other end sits the ambition of general AI: systems that can reason and learn across any domain the way a human can. Almost everything you encounter today falls firmly in the narrow category, even when it feels remarkably capable. Today's most capable AI products — the assistants, image generators, and coding tools — are genuinely impressive, but they remain narrow specialists: exceptionally good at language or images, not capable of the full range of human thought.
What has changed dramatically in recent years is not the idea of AI — researchers have been working on it since the 1950s — but how capable these systems have become, how cheaply they can be run, and how broadly they can be applied. The combination of vastly more data, much more powerful hardware, and a breakthrough architecture called the transformer has pushed AI out of research labs and into everyday products. That shift is still accelerating, which is why understanding the basics of AI has gone from a niche interest to genuinely useful knowledge for almost everyone.
Real-world example
When your phone's map app tells you the quickest way home and warns that your usual route is jammed, an AI system is behind it. It was never handed a rulebook of how traffic behaves on every street. It learned the patterns from millions of past journeys — how roads flow at different times of day, in different weather, on different days of the week — and applies what it learned to the trip you are taking right now. That is AI doing what it does best: learning from examples and applying that learning to a situation it has never seen in exactly that form before.
Related terms
Suggested courses
Curated pointers to take this further — these are suggestions, not reviews of courses we've completed.
Frequently asked questions
What is the difference between AI and machine learning?
Machine learning is one way of building AI — the most dominant approach today. It works by training systems on data rather than writing rules by hand. AI is the broader goal of making machines behave intelligently. Not all AI uses machine learning — early AI systems were built entirely on hand-crafted rules — but most modern AI does.
Is AI the same as automation?
Not quite. Automation follows fixed rules to handle predictable, repetitive tasks — a payroll system running at month end is automation. AI adds the ability to handle situations that weren't explicitly programmed, to learn from new data, and to make judgments under uncertainty. The two often overlap in practice — many AI-powered products also automate tasks — but the concepts are distinct.
Will AI replace human jobs?
Some roles will shrink, others will change, and some new ones will appear — which is broadly what has happened with every major technology shift. AI is currently better than humans at specific, well-defined tasks like pattern recognition in large datasets. It is still weak at complex reasoning, physical dexterity, original thought, and tasks requiring deep contextual judgment. The honest answer is that the full impact is still unfolding and anyone claiming certainty in either direction is overstating what we know.