Understanding Large Language Models Without the Jargon
A structured deep-dive into large language models — how they work, where they struggle, and what that means for people building with them today.
What this course covers
You have probably used ChatGPT, Copilot, or something similar at work. But knowing how to use a tool and knowing how it actually behaves are two different things.
This webinar is for people in roles like product management, content strategy, operations, or research who interact with LLM-powered tools regularly but did not come up through machine learning. The goal is not to make you a developer. It is to give you enough of a mental model that you stop being surprised by the weird things these systems do.
What makes LLMs behave the way they do
We will explain, without math, how language models learn from text, why they can write fluently about things they are completely wrong about, and what it means when a model has a knowledge cutoff. These are not edge cases. They come up in real workflows, and understanding them changes how you write prompts, review outputs, and set expectations with your team.
A significant part of the session focuses on practical judgment calls. When is an LLM a reasonable tool for a task, and when is it a liability? We will go through scenarios in content generation, summarization, data extraction, and customer communication, looking at where errors are low-stakes and where they are not.
Topics covered in this session
- How LLMs generate text and why they sometimes make things up
- Knowledge cutoffs, training data, and what the model does not know
- Prompt structure and why small wording changes produce different results
- Recognizing unreliable outputs before they cause problems
- Communicating AI limitations clearly to stakeholders
One attendee from a previous session described it as the first time AI stopped feeling like a black box and started feeling like a tool with known edges.
No prior technical background is assumed. If you can use a spreadsheet and write a professional email, you have everything you need to follow along.
Course programme
Webinar Program
Opening (10 min)
Context on why LLM literacy matters for non-technical roles right now, with a quick poll on how attendees currently use AI tools at work.
Core Concepts (40 min)
- How text generation works in plain language
- Hallucination explained with real examples from common use cases
- What training data and knowledge cutoffs mean in practice
Applied Judgment (35 min)
- Scenario walkthroughs: good use cases vs risky ones
- Prompt structure basics with live examples
- Spotting red flags in AI-generated content
Team Communication (15 min)
How to talk about AI limitations with colleagues and stakeholders without overstating or underselling what these tools can do.
Q&A (20 min)
Open questions with time to discuss specific tools or workflows attendees are using.
Materials included
Session recording, a one-page reference sheet on prompt writing, and a decision guide for evaluating whether an LLM is appropriate for a given task.
Questions before enrolling?
Send us your details and a brief note about your background. Our team typically responds within one business day and can help you decide if this course fits your current level.