AI Quick Tips 214: What is ChatGPT?

Since I think that there are more beginners out there than anything, this one is for you.

I’ve been thinking about a concise and accurate answer to this question for a while and it is harder than it seems.  All of the accurate answers would put you to sleep (aka - not concise) and all of the concise answers only cover a small fraction of what can be done.  

I think that a layered approach would be the best of both worlds.  Here’s what I mean -

What is ChatGPT? (Didn’t we ask this already…)

ChatGPT is a chatbot that gives you access to AI models

Great but what’s a chatbot and what are models?

A chatbot is a program that simulates human conversation.

According to IBM, an AI model is: 

A program that has been trained on a set of data to recognize certain patterns or make certain decisions without further human intervention

What kind of models can ChatGPT access?

• Large Language Models (LLMs) - simply put, they are text generation machines.  They take input, process it, and generate output
• Vision Language Models (VLMs) - models that can understand and process text and visual data (i.e., images, video, etc.)
• Multimodal models - models that can process information from text, images, video, audio, etc. (multiple modalities)

Put it all together

ChatGPT and other similar AI tools like Gemini, Perplexity, Claude and Copilot (among others) provide easy-to-use interfaces that allow users to access AI models.  Once people have access to the models they can do stuff.

What kind of stuff?  (Look at the rest of the blog…)

Now go experiment!