Artificial Intelligence (AI)
- What it is: AI is the overarching field of science dedicated to making computers and machines perform tasks that normally require human intelligence. This includes things like understanding language, recognizing images, solving problems, and making decisions.
- Analogy: Think of AI as the entire field of "transit” Just as "transit" is a huge category that includes specific modes like the bus, tram, car, bicycle, train. AI is a huge category that includes specific methods and applications like machine learning, computer vision, and robotics. It's a discipline, not a single technique.
- Relationship to the other terms: AI is the umbrella that all the other terms on this list fall under.
Machine Learning
- What it is: Machine Learning (ML) is the dominant sub-field of AI today. Instead of being explicitly programmed with a set of "if-then" rules, an ML system is "trained" by being shown huge amounts of data. It learns to recognize patterns within that data and then uses those learned patterns to make predictions or decisions about new, previously unseen data.
- Analogy: It’s like teaching a child to recognize a bus. You don't hand them a manual that says "if it has a metal skeleton, many wheels, and carries people around making regular stops, it’s a bus". Instead, you show them hundreds of pictures of different buses (you may have done this yourself by solving CAPCHAs on websites!). Their brain learns the "pattern" of what a bus looks like. ML works the same way, you feed a system thousands of labeled examples, and it builds its own model for identifying patterns.
- Relationship to the other terms: ML is a subfield of AI. It's the engine that powers most modern AI applications, including Large Language Models (LLMs).
Large Language Model (LLM)
- What it is: An LLM is a highly specialized type of machine learning model. It has been trained on an enormous amount of text (billions or trillions of words from books, articles, code, and the public web). Its core function is to predict the next word in a sequence. By doing this at a massive scale, it develops a sophisticated understanding of grammar, facts, reasoning styles, and context, allowing it to generate human-like text, answer questions, summarize documents, and even write code.
- Analogy: Think of it as "super-mega-autocomplete." Your phone often tries to guess the next word you'll type based on common phrases. An LLM does the same thing, but it's been trained on trillions of words and can predict the next word with an incredible understanding of context. The result is the ability to write a complete, coherent essay, not just finish your sentence.
- Relationship to the other terms: An LLM is a specific product of Machine Learning, which is a part of the broader field of AI. It is the technology behind systems like ChatGPT.
AI Agents (or "Agents")
- What it is: An Agent is an AI system that can perceive its environment, make plans, and take autonomous actions to achieve a specific goal. Modern AI Agents often use an LLM as their “brain” for reasoning, combined with tools like APIs, databases, or web browsers that allow them to take real actions.
- Analogy:
- Think of an LLM as a researcher that’s great at answering questions, and an Agent as an employee you can delegate tasks to.
- You can ask the researcher (LLM): "What are the cheapest trains to Chicago next Tuesday?" It will analyze the data it was trained on and give you a text-based answer.
- You can tell the employee (Agent): "Book me the cheapest train to Chicago next Tuesday." The agent will use its LLM "brain" to understand the goal, then use its "tools" (like a web browser to check train sites, a calendar API to check your schedule, and a booking API to purchase the ticket) to actually complete the task from start to finish.
- Relationship to the other terms: An Agent is an application of AI that often uses an LLM as its reasoning engine or "brain" to understand goals and decide which actions to take.
API
BONUS: An “API” (Application Programming Interface) is one of the "tools" that an AI Agent uses to perform actions and complete tasks. It’s a structured way for one piece of software to talk to another. It defines how programs can send requests, exchange data, and trigger specific actions without needing to know the other program’s internal code.
Analogy: Think of an API as the messenger that lets systems “talk.”
-
- A calendar API lets an AI Agent check your availability.
- A booking API lets it purchase a train ticket.
- A traffic API lets it check for delays.
The API doesn’t do the work itself, it’s the standardized connection that allows two systems to communicate safely.