AI - How do you use it?
When to Use AI (Best Practices)
- Brainstorming & Drafting: Use for creative ideas, generating outlines, or overcoming blank-page syndrome.
- Summarizing & Analysis: Use for rapidly digesting long documents or extracting data insights.
- Routine Tasks: Automate repetitive, low-stakes tasks to free up time for complex work.
- Learning & Coding: Use as a tutor or to troubleshoot tech problems and generate code snippets.
When to Avoid AI (Risks)
- High-Stakes Decisions: Avoid relying on AI for, legal, medical, or financial advice without expert human oversight.
- Deep Creative Work: AI lacks true innovation and human experience, often producing formulaic output.
- Deep Learning Tasks: Using AI to skip difficult, necessary learning processes can weaken your own skills.
- Sensitive Data: Be careful with confidential data, as some models may use your inputs for training.
Tips for Effective Usage
- Act as Curator: Treat AI as a junior assistant; you must check, edit, and verify all output, as models can hallucinate.
- Use for Variance: Ask for multiple, vastly different solutions (e.g., "give me 15 ways to write this") to break out of limited thinking.
- Select the Right Tool: Use more powerful models (like Claude 3.5 Sonnet, GPT-4o, or Gemini 1.5 Pro) for serious work.