AI series · Part 1: GenAI and Knowledge Modeling
Generative AI in Plain Words: What LLMs Can and Can't Do
Generative AI creates text, images and code by predicting the next word. Here's what that means, and why your company's knowledge is the part it's missing.
Generative AI is AI that creates new content (text, images, code) instead of only sorting things into categories. Large language models (LLMs) such as GPT, Claude, Gemini and Llama build an answer by predicting the next word, over and over. Think of your phone keyboard's autocomplete, but trained on most of the internet.
A simple example
You ask: "Write a leave request email." The LLM writes the email, with a greeting, dates, a reason and a polite sign-off. Nobody wrote that exact email before. The model generated it.
The key limitation: it only knows what it was trained on
An LLM has never seen your company's private documents, and it doesn't know what changed last week. Worse, when it doesn't know, it may hallucinate: it confidently makes something up.
| Question: "How many sick leaves do I get at Acme?" | Answer |
|---|---|
| A general chatbot | Guesses "10–12 days" from what is common elsewhere |
| An assistant grounded in Acme's policy | "12 days" [leave_policy#sick-leave], read from the real document |
The second answer is correct and shows where it came from. Getting there is what the rest of this series is about: retrieval (RAG), knowledge graphs and tools.
LLMs are strong at language but lack private, up-to-date knowledge, so in enterprise use we ground them with retrieval, knowledge graphs and tools.
- The short version
Next in the GenAI series: Taxonomy: Organize Knowledge So AI Searches Less and Finds More
Start from the top: GenAI Quick Answers and Series Guide
We deploy these patterns across production engagements in fintech, healthcare, and eCommerce. Talk to a senior engineer about how they apply to your platform.