This page is the parent hub for 3RK’s plain-English AI basics library. Use it when you know you need foundational context before jumping into tools, workflows, or safety decisions, but you do not know which explainer to read first.
Direct Answer
Most beginners do not need more hype. They need better vocabulary. Once you can separate models from chat products, prompts from fine-tuning, retrieval from training, and guardrails from full human judgment, the rest of the AI landscape becomes much easier to read.
This hub organizes the Basics category around that goal. Start with the question you actually have, then move into the child page that gives the clearest next step.
Evaluation Criteria
- The page helps readers find the right beginner explainer fast.
- The groups follow how people ask questions, not just technical taxonomy.
- The hub connects basics to real tool and workflow decisions.
- Safety and review concepts stay visible instead of becoming an afterthought.
- The page stays useful even if the reader knows only one or two AI terms today.
Library Overview
| Cluster | Best for | Main questions | Best first page |
|---|---|---|---|
| Core foundations | Readers who want the basic map first | What is AI? What is an LLM? How do these terms differ? | What Is AI? |
| Models and product layers | Readers confused by product language | What is a model? Is ChatGPT a chatbot or an assistant? What is an agent? | AI Models vs Chatbots vs Assistants vs Agents |
| Context, retrieval, and search | Readers hearing RAG, embeddings, tokens, and context windows | How does AI search work? Why do limits and retrieval matter? | What Is RAG? |
| Safety and control | Readers who want to understand risk, review, and guardrails | What can go wrong? What do guardrails do? Why is human review still needed? | AI Risk Terms Explained |
Where to Start by Question
| If your question is… | Start here | Then read | Why |
|---|---|---|---|
| What is AI, really? | What Is AI? | AI vs Machine Learning vs Generative AI | Start broad, then separate the overlapping labels. |
| What is an LLM and what makes it different? | What Is an LLM? | What Is Multimodal AI? | LLMs explain the core text model layer before you add other modalities. |
| Why do people keep saying RAG, embeddings, and context window? | What Is RAG? | Embeddings Explained | RAG is often the easiest entry point into retrieval vocabulary. |
| Why do prompts, limits, and costs behave strangely? | AI Tokens Explained | What Is a Context Window? | These two concepts explain many everyday AI frustrations. |
| What changes a model: prompting, fine-tuning, or training? | Training vs Fine-Tuning vs Prompting | Prompt Engineering Basics | Clarify the levers first, then improve how you use the lightest one. |
| How should I think about AI safety and human review? | AI Guardrails Explained | AI Risk Terms Explained | Guardrails and risk terms work best together. |
Core Foundations
- What Is AI? — the broad starting point for readers who need the simplest definition first.
- AI vs Machine Learning vs Generative AI — the fastest way to untangle the umbrella terms.
- What Is an LLM? — a beginner guide to large language models and where they fit.
- Prompt Engineering Basics — how better instructions improve results without changing the model itself.
Models, Interfaces, and Product Layers
- AI Models vs Chatbots vs Assistants vs Agents — a clean layer-by-layer explanation of the terms people keep mixing together.
- What Is Multimodal AI? — how AI works across text, images, audio, and real-world product workflows.
- What Is Inference in AI? — the live step behind model answers, speed, and scale.
- Open-Source vs Closed AI Models — what regular users should understand about control, hosting, and convenience.
- What Are AI Agents? — what changes when AI systems use tools, knowledge, and multi-step logic.
Context, Retrieval, and Search Concepts
- What Is RAG? — retrieval-augmented generation as a grounding pattern for source-backed answers.
- Embeddings Explained — how meaning-based search works behind the scenes.
- AI Tokens Explained — why prompts, outputs, limits, and costs often come back to tokens.
- What Is a Context Window? — the model’s working space for the current request.
- Training vs Fine-Tuning vs Prompting — how context and behavior levers differ.
Safety, Risk, and Review Concepts
- AI Risk Terms Explained — bias, hallucination, privacy, copyright, and safety in plain English.
- AI Guardrails Explained — what controls do, what they do not do, and where human review stays necessary.
- What Are AI Agents? — useful here again because tool use and multi-step action raise the stakes.
How to Use This Library
Start with one question, not the whole category. Read the article that best matches the confusion you have today, then follow the links outward into the more practical tool or workflow page only after the basic concept is clear.
If you are already choosing software, go next to the AI Tools Directory or the AI Tool Selection Matrix. If your concern is safer output or higher-stakes review, move into the Source Verification Checklist or the Human-in-the-Loop AI Automation Guide.
Review Checklist
- Pick the Basics page that matches the question you already have instead of reading the whole category at random.
- Use the context, retrieval, and search pages together when the confusion is about RAG, tokens, or long prompts.
- Use the safety pages when the job involves approvals, policy, higher-stakes output, or tool use.
- Move to tools and workflows only after the vocabulary is clear enough to make a better decision.
- Return to this hub when a new AI term starts appearing repeatedly across products or articles.
FAQ
Do I need to read every article in order?
No. This hub is meant to help readers choose the right starting page, not force a strict sequence.
What should I read first if I feel completely lost?
Start with What Is AI?, then AI vs Machine Learning vs Generative AI, and then What Is an LLM?.
What if my question is really about tools, not concepts?
Go next to the AI Tools Directory or the AI Tool Selection Matrix after reading the relevant Basics page.
Why are safety and review topics inside a Basics hub?
Because many beginners first encounter AI through products that feel simple, even when the risk and review questions are not simple at all.
Verified External Sources
- OpenAI key concepts
- OpenAI prompt engineering guide
- OpenAI embeddings guide
- OpenAI safety in building agents
- Google Cloud: What is AI inference?
- Anthropic context windows guide
- Google Gemma model overview
- Anthropic mitigate jailbreaks and prompt injections