Module A opener matrix · Foundational tier · Read this first to see the tools clearly, then we'll talk about how to actually use them.

Summary

Before you start doing office work with AI, get one thing straight: there are basically two kinds of tools — the general assistant and the dedicated office AI. Don't use them at random, and don't let the long list of "AI tool" names scramble your brain.

The Pain Point: Sound Familiar?

Lots of people get stuck at two opposite extremes when using AI for office work:

  • Only knows how to "chat" with ChatGPT. Great at asking it "where should I go this weekend" or "how do I comfort a friend who just got dumped," but the moment it's time to get real work done — write a weekly report, fix an Excel sheet, make a PPT — they have no idea where to start. After the chat, they still have to do everything from scratch themselves.
  • Confused by the tool names. Today you see "Copilot," tomorrow you hear "WPS AI," and the day after a "ERNIE Bot" or "Qwen" pops up. A pile of names, no clue who does what — so you either force everything through a chatbot, or buy a bunch of memberships you never use.

Here's the truth that cuts through all the noise: almost every AI office tool on the market falls into just two categories. Get these two straight, and you'll stop getting lost.

Type 1: The General Assistant

Represented by: ChatGPT, Claude, Gemini, Grok — products everyone calls "large-model chat assistants."

In plain language: it's like an all-purpose assistant who can talk about anything — you type a question, it types back an answer, and it thinks lightning-fast.

What it can do:

  • Brainstorm with you and spark ideas ("How do I make this proposal more persuasive?")
  • Write copy and polish wording ("Help me draft a product intro")
  • Work as a search engine ("How do I remove duplicates in Excel?")
  • Help you build a framework and outline your points

The key limitation: it does not touch your documents directly. You say "polish my Word doc," and it hands you a fresh block of text that you have to copy back yourself. You say "add up this table," and it gives you a formula and steps that you have to perform yourself in the software. It's the strategist who "gives ideas and materials" — it never steps into your workspace.

Type 2: The Dedicated Office AI

Represented by: Microsoft 365 Copilot, WPS AI, Tencent Docs AI.

In plain language: unlike a chatbot that lives on its own, it "lives inside" your Word, Excel, and PPT — while you're writing the document, it sits right beside you and edits it directly.

What it can do:

  • In Word, directly expand, polish, and summarize this document for you
  • In Excel, directly clean your data, write formulas, and build charts
  • In PPT, directly lay out pages, add visuals, and generate slides

The key advantage: it edits your file directly. You say "make this paragraph more formal," and it changes the words in your document; you say "make a chart from this table," and it literally draws it inside your sheet. It's the apprentice who "does the hands-on work" — right there in your software.

Picking the Right Tool: One Line to Remember

Brainstorming, writing, shaping ideas → use a general assistant; editing documents, building tables, making PPTs → use the dedicated software.

Two small examples make the difference obvious:

  • ✅ Let ChatGPT draft your weekly report: you tell it "write this week's operations report; focus on user acquisition and events," and it returns a finished piece of text — that's "producing content," which a general assistant handles best.
  • ✅ Use WPS AI to polish this document directly: you open the report, select a paragraph, click "AI Polish," and it rewrites the sentences on the spot — that's "editing a file," which a dedicated office AI does most painlessly.

Don't run them backwards, either: you ask ChatGPT to "fix my Excel," and it can't — it can only teach you how; you ask WPS AI to "come up with a startup idea out of thin air," and it's less imaginative than a general assistant. Each does its own job, and that's the most efficient.

Advanced Track

Now that "general vs dedicated" is clear, the next step is to get your hands dirty. Module B is the hands-on section — pick the scenario that fits you:

  • B1 Expense Sheet: use AI to build a clean, standard expense sheet → line items, formulas, auto-totals
  • B5 Work Summary: let AI turn scattered tasks into a presentable summary
  • B9 Text to PPT: turn a block of text straight into slide after slide
  • B13 Access Table: use AI to build a lightweight database table and manage structured data
  • B17 PowerShell: have AI write scripts that automate those repetitive chores in batches
  • B21 Power BI: use AI to turn data into a dashboard you can read at a glance
  • B25 Python: use AI to write Python for the heavy lifting Excel can't handle

(Above are planned internal links; once each article goes live, they'll link directly.)

Next Steps

If you want AI to actually follow your orders, the most critical thing is how you write your prompt (a "prompt" is just the instruction you give the AI).

👉 A3 "Prompting 101: Make AI Understand Your Task" — next up, it teaches you how to turn "help me write a weekly report" into a clear instruction AI can really deliver on.

This article is part of vba.net's "AI-Powered Office: From Beginner to Expert" series, foundational tier. Author: He Xun. Translator: Yi An. Please credit the source if reprinted.