10 min · up to 50 XP

The model landscape

ChatGPT, Claude, Gemini, Copilot: what's actually different, and which one to reach for when.

Word of Priya's AI-assisted board summary got around. By Friday, three colleagues had cornered her at the coffee machine with the same question: "which one should I use?" One swore by ChatGPT. Another had Copilot "because it's already in Outlook". A third had read Claude was better for writing; someone in finance backed Gemini. Everyone was certain. Nobody could say why.

So before we name any strengths, you decide. Priya's next task is real. Which tool would you reach for?

Tap the diagram

Priya's just been handed a 38-page supplier contract to dissect before Friday. Tap the tool you'd reach for.

Match the tool to the task, not the headlines.

Meet the four properly. ChatGPT (OpenAI): the one that started the boom, most widely used, biggest ecosystem. Claude (Anthropic): a reputation for careful, natural writing and very long documents. Gemini (Google): woven through Gmail, Docs, Sheets and Drive. Copilot (Microsoft): AI threaded through Word, Excel, Outlook and Teams.

And the crucial part: every one of them is an LLM. The lesson-one mental model (prediction engine, no fact database, capable of hallucinating) applies to all four, in full. Switching tools never switches off the rules.

Fill the gaps

Fix the mental map before we go tool-shopping:

ChatGPT, Claude, Gemini and Copilot are all , so all of them can . Skills learned on one .

The difference people miss isn't between brands; it's within each one: model tiers. Every provider offers a spectrum, from fast-and-cheap models for everyday drafting and quick rewrites, to deep-and-capable ones for careful analysis, tricky reasoning and long documents.

This is where free vs paid usually bites: free plans typically give you the faster tiers with usage limits; paid plans unlock the deeper models. For a quick email, free is genuinely fine. For that 38-page contract, the tier matters more than the logo. Match the tier to the stakes, not the brand to the hype.

Predict the response

Priya's on a free plan's fast tier, the everyday model. She pastes in the long, dense supplier contract and asks for a careful risk analysis. What comes back?

Here's our 38-page supplier contract. Give me a careful analysis of the clauses that put us at risk.

What do you think the AI will do?

So when does the brand matter? When the tool sits closer to where the work already is. Company lives in Microsoft 365 or Google Workspace? Copilot or Gemini can already see your emails, documents and calendar: no endless pasting. Long documents and careful prose? Claude, as you found. Biggest ecosystem and the most shared guides? ChatGPT.

Notice these are workflow reasons, not "which AI is smartest" reasons. The honest answer to "smartest" changes every few months, which is exactly why you shouldn't build your habits on it.

Build the prompt

Priya's company runs on Microsoft 365. Leadership wants a one-page summary of Monday's product launch meeting; the notes are sitting in her Teams and Outlook. Choose the right tool, then the right ask.

1 · Which tool does she reach for?

2 · What does she ask for?

Tip

Pick one tool and go deep. Fluency with one assistant beats a shallow acquaintance with four; the skills transfer, because it's the same technology underneath. Then spend twenty minutes each quarter scanning what's changed: a quarterly glance keeps you current without turning tool-watching into a hobby.

Key takeaways

The big four (ChatGPT, Claude, Gemini, Copilot) are all LLMs: one mental model, one set of strengths and failure modes, everywhere.
You chose by task, not hype: Claude for the long contract; Copilot/Gemini where your office suite lives; ChatGPT for ecosystem and ubiquity.
You watched a fast tier skim that contract, fluent but shallow. Tiers within a tool matter more than the logo: match the tier to the stakes.
Pick one, go deep, revisit quarterly: your skills transfer, and the landscape will have shifted again by the time you look up.