Machine learning
The part of AI where the software works out the rules itself. Instead of a person writing the logic, the system is shown examples and finds the pattern that fits them, then applies it to cases it has never seen. Everything downstream is a variation on that: which examples you supply, whether anyone has labelled them, and how the system is scored when it gets one wrong.
Artificial Narrow Intelligence, also called weak AI. One job, one domain, usually done better than a person does it. Search algorithms, video recommendations, drones, invoice recognition, the systems we build.
ChatGPT belongs on that list too, which surprises people, because it will discuss anything you raise. But it is doing one job the whole time: take the text so far and continue it. Law, medicine, poetry are the same task to it, with different words in front. The subject matter is wide. The capability is not, and narrow AI is the only kind that exists.
AI doesn’t think. AI computes.
…mathematical formulas turning inputs into predictions, classifications or decisions. Machine learning chains them into systems that find patterns in data on their own. The simplest way to hunt a pattern is to break information into elementary logical paths, a sequence of…