The assistant landscape — one map, no favourites

Walk into any phone shop and every handset looks different until you notice they all make calls, all run apps, all take photos — and then the real differences start to matter. AI assistants are the same. Module E1 gave you the engine; this module walks the showroom.

Every major assistant — ChatGPT, Claude, Gemini, Copilot, Perplexity, Grok, Meta AI, and the rest — is the same machine you already understand: a large language model (E1·L6) wrapped in a chat interface, increasingly with tools attached (web search, file reading, code running, image generation). Your E1 skills — the three-question habit, the capability map — transfer to all of them, and to every assistant launched after this lesson was written. That’s why we taught concepts first.

Where they genuinely differ

Model quality per task. Vendors ship different model tiers; strengths shift with every release. Today’s best writer may not be next quarter’s.
Attached tools and integrations. Which assistant can search the web, read your files, sit inside your email or documents?
Limits and pricing. Free tiers differ in daily usage, model access, and features — and change often, which is why our tool cards carry a “verified” date instead of pretending permanence.
Data handling. What is stored, used for training, or excludable varies — and matters at work (E8 covers policy).

How this course stays neutral

No vendor pays us; no tool is “ours”. Each assistant lesson wraps a registry card — a dated, community-verified fact sheet — plus guidance on what tasks each tool suits. When a card ages past 120 days unverified, you’ll see a warning banner rather than stale confidence. Where a tool is weak, the lesson says so.

Try it now (2 minutes)

Write down which assistant(s) you already use, and one thing that annoys you about each. Keep the note — by the end of E2 you’ll know whether the fix is a technique (usually) or a different tool (sometimes).

Check your understanding

1. Why do your E1 skills transfer to every assistant in this module? All assistants share the same LLM machinery underneath The assistants copy each other's interfaces They don't — each tool needs relearning from zero

2. An assistant comparison you read is eight months old. Best attitude? Treat it as current — these tools rarely change Treat specifics as expired: model tiers, limits, and features shift fast Assume the opposite is now true

Recap

Same engine everywhere; real differences on four axes; dated facts over brand loyalty. Next: the assistant most people meet first — ChatGPT.