Sign In
Signed out 5
Module 05 · Tier 2 — Basic Prompting

AI Tools Landscape

ChatGPT, Claude, Gemini, Copilot — what each is good for and how to choose.

10 min Tier 2 — Basic Prompting
Why This Matters

The AI tools landscape changes fast. But the principles for choosing the right tool for a given task change much more slowly. This lesson gives you a framework for evaluating AI tools — now and as new ones emerge — plus an honest look at where today's leading tools have different strengths.

The Concept

The Four Categories of AI Tools

Most AI tools you'll encounter fall into four categories, each suited to different types of tasks:

1. General-Purpose Chat (Claude, ChatGPT, Gemini)

Conversational interfaces for writing, analysis, explanation, brainstorming, and research synthesis. These are the Swiss Army knives of AI — not always best-in-class for specific tasks, but remarkably capable across a huge range of uses.

Use when: you need to think through a problem, generate text, analyze a document, explain a concept, or aren't sure which specialized tool fits.

2. Code-Augmented AI (GitHub Copilot, Cursor, Replit)

AI integrated into development environments, with context awareness of your specific codebase. Better than general chat for coding tasks because they can see your actual files, error messages, and project structure.

Use when: you're writing, debugging, or refactoring code in an active project.

3. Search-Augmented AI (Perplexity, Bing Copilot, Google AI Overviews)

AI that can retrieve current information from the web and cite sources. Reduces hallucination risk for factual queries because it can access real-time data. Still not a substitute for reading primary sources on important decisions.

Use when: you need current information, research synthesis with citations, or factual queries where training data cutoffs matter.

4. Workflow-Integrated AI (Microsoft Copilot for M365, Notion AI, Slack AI)

AI built into the tools you already use, with access to your organizational data — emails, documents, meetings. Context-rich because it knows your actual work, not just general knowledge.

Use when: you're working inside those platforms and need AI that has context on your specific work.

Choosing Between General-Purpose Tools

For the three leading general-purpose tools as of early 2026:

Claude (Anthropic): Particularly strong at following nuanced instructions, long-document analysis, writing with a specific voice, and careful reasoning through complex problems. Tends toward being thorough and honest about uncertainty. Good default for writing, analysis, and anything where you need careful reasoning.

ChatGPT (OpenAI): Largest ecosystem of plugins and integrations. Strong code generation, image generation (with DALL-E), and broad general capability. Wide range of specialized GPTs for specific use cases. Good default when you need tool integrations or specialized functionality.

Gemini (Google): Best integration with Google Workspace (Docs, Gmail, Drive). Strong multimodal capability (images, audio, video). Native access to Google Search grounding. Good default if you live in Google's ecosystem.

Practical advice: Don't over-optimize your tool choice. The difference between using the "best" tool and using any of these three well is much smaller than the difference between prompting poorly and prompting effectively. Develop skill with one tool before shopping for another.

The Evaluation Framework for New Tools

New AI tools launch constantly. When evaluating a new one, ask:

  • What specific problem does this solve better than general-purpose AI?
  • What data does it have access to that general AI doesn't?
  • What's the privacy and security model for data I put into it?
  • Is the improvement worth the switching cost and learning curve?
Side-by-side: the same prompt in different tools

One valuable experiment: run the same prompt in two different tools and compare the responses. The differences will often reveal the different strengths, defaults, and tendencies of each model.

Try: "Summarize the main arguments for and against using AI for hiring decisions. Be balanced." Compare the depth, the caveats, the tone, and what each model emphasizes or omits.

Hands-On Exercise

Map your use cases to tools

ClaudeChatGPTGeminiPerplexity
List the three most common tasks you currently use AI for (or would like to use AI for). For each one, think through: 1. Which tool category does this task fall into? 2. Which specific tool might be best suited to it? 3. Is there a privacy or data consideration that affects which tool you should use? Then: try the task in the tool you identified. Was the fit as good as you expected?
If you work with sensitive data, the privacy question is not optional — it's the first question.
Active Recall

Before moving on — close this lesson and answer these from memory. Then come back and check. Testing yourself (not re-reading) is how this sticks.

1 What are the four categories of AI tools, and what distinguishes each from the others?
2 When would you choose a search-augmented AI tool over a general-purpose chat tool?
Reflection

Which tool or category of AI tools are you currently underusing? Based on this lesson, what would you try differently this week?

Key Takeaway

Match the tool to the task. General-purpose chat for writing and analysis. Code-augmented for development. Search-augmented for current facts. Workflow-integrated for organizational context. Skill with any tool beats using the "right" tool poorly.