Module 2: Research Tools & Models
AI for Research & Analysis
This module helps you pick the right tool for the job.
The Major Players
ChatGPT (OpenAI)
- Best for: General research, explanations, brainstorming
- Strengths: Large knowledge base, good at conversation, widely used
- Weaknesses: Knowledge cutoff, can hallucinate confidently
- Web browsing: Available (Plus/Team plans)
Claude (Anthropic)
- Best for: Long documents, nuanced analysis, detailed explanations
- Strengths: Handles large context (100k+ tokens), thoughtful responses
- Weaknesses: Smaller ecosystem, knowledge cutoff
- Web browsing: Not built-in
Perplexity AI
- Best for: Fact-finding with sources, current information
- Strengths: Cites sources, real-time web access, focused answers
- Weaknesses: Less good at synthesis, shallower analysis
- Web browsing: Core feature
Google Gemini
- Best for: Multimodal research (images, documents)
- Strengths: Google integration, handles multiple file types
- Weaknesses: Variable quality, still maturing
- Web browsing: Built into Google ecosystem
Elicit
- Best for: Academic research, finding papers
- Strengths: Searches academic databases, extracts findings
- Weaknesses: Academic focus only, limited scope
- Web browsing: Academic databases
When to Use What
| Research Type | Best Tool | Why |
|--------------|-----------|-----|
| Quick fact check | Perplexity | Real-time sources, citations |
| Deep topic learning | Claude/ChatGPT | Conversation, explanations |
| Academic research | Elicit | Paper-focused, structured |
| Current events | Perplexity + Web | Real-time access |
| Long document analysis | Claude | Large context window |
| Visual/image research | Gemini | Multimodal capabilities |
| General starting point | ChatGPT | Versatile, reliable |
Match each AI tool to its research strength:
The Multi-Tool Approach
Smart researchers don't pick one tool — they use the right tool for each phase:
Phase 1: Orientation
ChatGPT or Claude to understand the landscape
Phase 2: Source Finding
Perplexity or Elicit to find actual sources with citations
Phase 3: Deep Reading
Claude for long documents, ChatGPT for Q&A
Phase 4: Synthesis
Claude or ChatGPT to bring it all together
Free vs. Paid: What You Actually Need
Free tiers get you:
- Basic research capabilities
- Limited message/query counts
- Smaller context windows
- Older models
Paid tiers add:
- Latest models (usually much better)
- Larger context windows
- Web browsing
- File uploads
- More messages
Worth paying for research?
If you research regularly: Yes. The productivity gain pays for itself.
If you research occasionally: Free tiers are fine. Use Perplexity for citations.
Put these research tool selection criteria in order of importance:
Model Versions Matter
Within each platform, newer models are usually better:
- GPT-4 >> GPT-3.5
- Claude 3 Opus >> Claude 3 Haiku
- Latest Gemini >> Older versions
For research, use the best model you have access to. The difference is significant.
Specialized Research Tools
Beyond the big AI assistants:
Consensus — Academic paper search with AI summaries
Semantic Scholar — AI-powered academic search
Connected Papers — Visual paper relationship mapping
Scite — Shows how papers cite each other (supporting/contrasting)
NotebookLM — Google's research notebook with AI
Each serves a specific research niche.
Building Your Research Stack
A practical setup:
1. Primary AI assistant — ChatGPT or Claude (for thinking partner)
2. Citation-focused tool — Perplexity (for sourced facts)
3. Academic tool — Elicit or Semantic Scholar (if you need papers)
4. Note-taking — Notion, Obsidian, or Roam (for organizing findings)
5. Traditional search — Google, Google Scholar (still essential)
You don't need everything. Start with one AI assistant and Perplexity.
Privacy Considerations
AI tools learn from conversations (usually). Consider:
- Don't paste confidential data into free AI tools
- Check data policies before uploading sensitive documents
- Enterprise plans typically offer better privacy protections
- Some tools offer "no training" modes
For sensitive research, read the privacy policy first.
---
Tool Selection Helper
I need to research [TOPIC] for [PURPOSE]. My constraints: - Budget: [FREE/PAID/SPECIFIC AMOUNT] - Time: [HOW LONG I HAVE] - Depth needed: [SURFACE/MODERATE/DEEP] - Source requirements: [NEED CITATIONS? ACADEMIC? CURRENT?] Recommend which AI tools I should use and in what order.
💡 Use this to plan your tool stack before starting.
Multi-Tool Research Plan
I'm researching [TOPIC]. Create a research plan that uses multiple AI tools: 1. What to use for initial orientation 2. What to use for finding sources 3. What to use for deep analysis 4. What to use for synthesis For each step, tell me: - Which tool and why - What to ask/upload - What output to capture
💡 Plan your workflow before jumping in.
Tool Capability Check
I want to use [AI TOOL] to research [TOPIC]. What can this tool do well for this research? What are its limitations I should know about? What should I use other tools for instead? Any specific tips for using this tool effectively?
💡 Understand your tool's strengths and limits.
1. Sign up for Perplexity (free tier is fine)
2. Run the same question through your primary AI (ChatGPT/Claude) and Perplexity
3. Compare: Which gave better sources? Which explained better?
4. Note which tool you'd use for what
You now have two complementary research tools.
---
- 1Different AI tools excel at different research tasks
- 2Use Perplexity for citations, ChatGPT/Claude for thinking
- 3Multi-tool workflows beat single-tool dependency
- 4Paid tiers are worth it for regular researchers
- 5Build a stack: primary AI + citation tool + notes system