AI Knowledge

The handful of ideas that explain almost everything you hear about AI.

Artificial intelligence

A broad term for software that performs tasks we usually associate with human thinking: understanding language, recognizing images, making decisions, predicting what comes next.

Machine learning

Instead of hand-writing rules, we show the computer many examples and let it find the patterns itself. A spam filter trained on a million labeled emails is machine learning.

Neural networks & deep learning

Neural networks are models loosely inspired by the brain: layers of simple units that transform inputs into outputs. “Deep” just means many layers. Given enough data and computing power, deep networks learn remarkably subtle patterns — this is what powers modern image, speech, and language systems.

Large language models (LLMs)

LLMs are neural networks trained on huge amounts of text. Their core skill is predicting the next word, but at scale this produces fluent writing, translation, summarization, reasoning, and code. They don’t “know” facts the way a database does — they generate responses from learned patterns, which is why they can occasionally be confidently wrong. Always verify important outputs.

Prompts & context

A prompt is your instruction to the model. Good prompts are specific about the task, the audience, the format you want, and any constraints. Context — background information you paste in — is what turns a generic answer into a useful one.

AI agents

An agent is an AI that doesn’t just answer — it acts. It can use tools (search the web, read your calendar, run code), remember your preferences, and carry out multi-step tasks on your behalf, checking in when a decision is genuinely yours to make.