Introduction to AI Models: Why Are There Different Ones?

Reading time: approx. 7 min

Welcome to the course "Learn the Most Important AI Models"! In this course we will explore the different AI models you may encounter and benefit from in your teaching. Before we dive into specific models like ChatGPT, Gemini and Copilot, it is important to understand what an AI model actually is and why there are so many different types.

What you will learn

What an AI model is at a fundamental level.
Why different AI models have different strengths and weaknesses.
Which main categories of AI models are relevant for teachers.

The Basics: What is an AI Model?

An AI model can be seen as an advanced computer program that has been trained on enormous amounts of data to perform specific tasks. Think of it as a student who has read thousands of books, articles and conversations to become an expert in a field. When you then ask a question or give a task to the AI model, it uses its trained knowledge to generate a relevant answer or perform the action.

The models we will focus on in this course are primarily large language models (LLMs). These models specialize in understanding, generating and processing human language. They can write texts, summarize information, answer questions and even translate. Some of them are also multimodal, which means they can handle not only text but also images, audio and video.

Why Are There Different AI Models?

Just as there are different types of teachers with different specialties, there are different AI models developed with different purposes, training data and architectures. Here are some of the main reasons for this diversity:

Different training data: Some models are trained more on text from the internet, others on specific datasets such as scientific articles, code or conversations. This affects what type of information they are best at handling and generating.
Focus and optimization: Developers optimize models for different purposes. One model may be optimized for speed, another for creativity, a third for factual precision and a fourth to be as small and efficient as possible on smaller devices.
Technical architecture: Under the hood there are different ways to build these models. Some are designed to be large and powerful, while others are more efficient with less computing power.
Commercial interests and values: Companies like OpenAI, Google and Anthropic have different philosophies and business models, which affects how they design and limit their models. Some prioritize openness and accessibility, while others focus more on safety and enterprise integration.
Multimodality: Some models are text-only, while newer models have the ability to understand and generate information across multiple modalities such as text, images and audio. This significantly expands their use cases.

Main Categories of AI Models Relevant for Teachers

For you as a teacher, we can roughly divide the most relevant AI models into the following categories:

General language models (LLMs): These are the most versatile. They can generate text, summarize, answer questions and assist with creative writing. Examples: ChatGPT (OpenAI), Claude (Anthropic), Gemini (Google).
- Some are multimodal: Within this category there are models that can also handle images, such as ChatGPT with GPT-4o and Gemini.
Integrated AI assistants: These models are often embedded in existing software and services to improve productivity and efficiency. Example: Microsoft Copilot (which uses underlying LLMs like GPT-4).
Specialized models (brief mention): Although we will not go into depth on these, there are models specialized in areas such as image generation (e.g. DALL-E which ChatGPT uses for image generation), code generation (like Claude Code), or research (like Perplexity AI). Grok is another example of a model with a specific niche and personality. There are also technology companies like Groq (not to be confused with Grok), which develop hardware to run AI models extremely fast, rather than being a model itself.

Practical Examples: Why Does This Matter to Me?

Understanding these differences is crucial for choosing the right tool for the right task:

Do you need to brainstorm ideas for a new lesson? A general LLM like ChatGPT or Gemini can be excellent.
Do you need to summarize a long article or policy to better understand it? Claude can be a strong choice thanks to its large context window.
Do you want to quickly create a presentation or draft an email directly in Microsoft Office? Microsoft Copilot is then your primary tool.
Do you need an AI that can generate images to illustrate a concept? Then you need a multimodal model like ChatGPT (with GPT-4o) or Gemini.

By knowing the specific strengths and limitations of AI models, you can maximize their potential and avoid frustration when they do not perform as expected.

Next Steps

In the next module we will dive into perhaps the most well-known AI model of all: ChatGPT. We will look at different versions, use cases and how it can become an invaluable assistant in your pedagogical everyday life.