ChatGPT Fundamentals

ChatGPT is a conversational AI product built by OpenAI. It launched in November 2022 and crossed 100 million users faster than any consumer product in history. But understanding what it actually is, rather than the hype around it, matters for getting real value from it.

At its core, ChatGPT is a text interface to a family of large language models. You type something, the model generates a response based on patterns it learned during training. It does not search the internet in real time by default (though it can with browsing enabled), it does not have persistent memory across conversations unless you turn that feature on, and it does not "think" the way humans do. It predicts the next likely sequence of text based on your input and its training data.

That sounds limiting, but the practical capability is enormous. ChatGPT can write essays, debug code, summarize documents, translate between languages, brainstorm ideas, roleplay as a tutor, analyze data, generate images, and carry on nuanced conversations about nearly any topic. The gap between "predicts text" and "genuinely useful tool" is smaller than most people expect.

The key mental model: ChatGPT is a highly capable collaborator that works best when you give it clear context, specific instructions, and feedback. It is not a magic oracle. The quality of what you get out is directly proportional to the quality of what you put in. This entire course is built around helping you put in better inputs and get dramatically better outputs.

OpenAI offers several models through ChatGPT, and knowing which one you are using matters more than most people realize.

GPT-4o is the default model for free and Plus users. The "o" stands for "omni" because it handles text, images, audio, and video natively. It is fast, capable, and covers 90% of everyday tasks well. For most users, GPT-4o is the right starting point and often the right finishing point too.

GPT-4.5 is OpenAI's largest and most knowledgeable model. It has significantly more world knowledge, writes more naturally, and handles nuanced topics with more depth. It is available on Plus and Pro plans. The tradeoff is that it is slower and more expensive in API usage. For creative writing, in-depth research, and tasks where quality matters more than speed, GPT-4.5 is a meaningful upgrade.

o-series models (o1, o3, o4-mini) are reasoning models. They work differently from GPT-4o and 4.5. Instead of generating a response immediately, they "think" through the problem step by step before answering. This makes them dramatically better at math, logic puzzles, complex coding challenges, and scientific reasoning. The tradeoff is speed: a reasoning model might take 30 seconds to answer a question that GPT-4o answers in 2 seconds. Use them when accuracy on hard problems matters more than speed.

The model picker in ChatGPT (the dropdown at the top of the chat) lets you switch between models mid-conversation. A practical workflow: start with GPT-4o for speed, then switch to a reasoning model if the task requires deeper analysis.

ChatGPT's pricing tiers are straightforward once you cut through the marketing language.

Free tier gives you access to GPT-4o with daily message limits that vary (typically generous enough for casual use), basic image generation with DALL-E, web browsing, file uploads, and access to the GPT Store. For someone exploring ChatGPT for the first time, the free tier is genuinely useful and not just a teaser. You can accomplish real work with it.

ChatGPT Plus ($20/month) removes most usage limits on GPT-4o, adds access to GPT-4.5 and o-series reasoning models, gives you higher priority during peak usage times, and unlocks advanced features like Canvas (a collaborative editing workspace) and Operator (an AI agent that can browse the web and take actions on your behalf). For professionals using ChatGPT daily, Plus is where the tool becomes genuinely powerful.

ChatGPT Pro ($200/month) is the unlimited tier. Unlimited access to all models including the most capable reasoning modes, unlimited Operator usage, and priority access to new features. Pro makes sense for power users who depend on ChatGPT for critical daily workflows and need guaranteed access to the best models without rate limits.

Team ($25/user/month) and Enterprise (custom pricing) add workspace collaboration, admin controls, and enhanced data privacy guarantees. These matter for businesses, not individuals.

The practical recommendation: start with free, upgrade to Plus when you hit the daily limits or need reasoning models, and only consider Pro if you are using ChatGPT as a core part of your professional workflow for several hours every day.

The ChatGPT interface is deceptively simple, but several features are easy to miss if nobody points them out.

The sidebar (left panel) stores your conversation history, organized by time. You can search through past conversations, rename them for easy reference, and organize them into folders. Get in the habit of using descriptive first messages so the auto-generated titles are useful. "Help me write a Q3 marketing report" is searchable later; "hey can you help me" is not.

The model picker (top of chat) lets you switch models. The small text below the model name tells you which capabilities are active (browsing, DALL-E, code interpreter, etc.).

Attachments (the paperclip icon) let you upload files: PDFs, images, spreadsheets, code files, and more. ChatGPT can read and analyze these directly. This is one of the most underused features. Instead of copying and pasting text from a document, just upload the whole file.

The @ mention feature lets you invoke Custom GPTs mid-conversation. Type @ and you will see your recently used GPTs. This is powerful for chaining different specialized tools without leaving your chat.

Canvas (available on Plus) opens a side-by-side editing workspace. When ChatGPT generates a long document or code file, Canvas lets you highlight specific sections and give targeted feedback. It is particularly useful for iterative writing and coding tasks.

Memory (Settings > Personalization > Memory) lets ChatGPT remember facts about you across conversations. When you tell it "I run a SaaS company targeting enterprise customers," it stores that and uses it in future conversations. You can review and delete individual memories at any time.

The difference between a bad prompt and a good prompt is often the difference between useless output and output you can actually use. The good news is that effective prompting follows a few simple principles.

Be specific about what you want. "Write about marketing" produces generic filler. "Write a 300-word LinkedIn post explaining why small businesses should invest in email marketing over social media, with a conversational tone and a specific call to action to download a free guide" produces something usable.

Provide context. ChatGPT does not know your audience, your industry, your constraints, or your preferences unless you tell it. The more relevant context you include, the better the output. "I am a freelance web developer writing a proposal for a local restaurant that wants a new website. Budget is $3,000. Write the proposal." That context shapes everything about the response.

Specify the format. Want bullet points? Say so. Want a table? Ask for it. Want the response structured as an email with subject line, greeting, body, and sign-off? Describe that structure. ChatGPT will follow formatting instructions precisely if you provide them.

Iterate, do not start over. Your first prompt rarely produces perfect output, and that is fine. The real skill is follow-up: "Make the opening stronger." "Cut this to half the length." "Add more specific examples from the restaurant industry." "Change the tone from formal to friendly." Each follow-up refines the output. Think of it as a conversation, not a single command.

Give examples when possible. If you have a writing sample that represents the tone you want, paste it in and say "Write in this style." If you have a template you want followed, include it. Few-shot prompting (giving examples of what you want) is one of the most effective techniques, and you will master it in Module 2.