Tips

When NOT to Use AI: Knowing the Limits

January 8, 2025 3 min read

AI is powerful. It’s also not always the right tool.

Here’s when you should step away from the AI and do it yourself.

When AI Output Quality Suffers

1. When You Know More Than It Does

You’re an expert. AI is a generalist.

If you’re a lawyer drafting a contract, a doctor explaining a diagnosis, or an engineer solving a specific problem — your expertise matters more than AI’s generic knowledge.

AI is a starting point, not the authority.

2. When Accuracy Is Critical

AI confidently states incorrect things.

Don’t use AI as your source of truth for:

  • Medical information
  • Legal advice
  • Financial decisions
  • Safety-critical information
  • Anything you’d normally verify

Verify everything that matters.

3. When Nuance Matters

AI flattens nuance.

Sensitive communications — delivering bad news, handling conflicts, navigating delicate relationships — require human judgment.

Some messages need a human touch AI can’t replicate.

When AI Creates More Work

4. When It’s Faster to Just Do It

Five minutes to write a prompt, wait for output, edit the result…

Or two minutes to just write the email yourself.

Don’t use AI to avoid 2-minute tasks.

5. When You’ll Rewrite Everything Anyway

If AI output requires complete rewriting, you saved nothing.

Sometimes starting from scratch is faster than fixing AI output.

If you’re editing more than creating, skip the AI step.

6. When Learning Is the Point

If you need to learn something, struggling with it is part of the learning.

Having AI do your homework doesn’t teach you anything.

Don’t shortcut learning you actually need.

When AI Is Ethically Problematic

7. When It’s Someone Else’s Test

Using AI for:

  • Job applications (representing it as your work)
  • Academic submissions (against policy)
  • Client work (without disclosure)
  • Anything misrepresenting its origin

If you’d be embarrassed to admit you used AI, don’t use AI.

8. When Human Judgment Is Owed

Some decisions deserve human consideration:

  • Hiring/firing decisions
  • Medical diagnoses
  • Legal judgments
  • Personal relationships

People deserve human judgment for high-stakes decisions.

9. When Privacy Is At Stake

Putting sensitive information into AI tools:

  • Personal data
  • Confidential business information
  • Private communications
  • Health information

Don’t feed AI information you wouldn’t share publicly.

When AI Undermines Your Goals

10. When Voice Matters

Your unique voice is valuable.

If readers follow you for YOUR perspective, AI-written content undermines that.

Generic AI content doesn’t build authentic connection.

11. When Creativity Is the Goal

AI remixes existing ideas. It doesn’t truly create.

For genuine creative work, AI might constrain rather than expand your thinking.

Don’t let AI’s suggestions limit your creative range.

12. When Relationships Are the Goal

Using AI to write all your personal messages creates distance.

The point of communication is connection, not efficiency.

Some messages should be inefficient and personal.

The Decision Framework

Before using AI, ask:

  1. Will I verify this? If you won’t verify, don’t use AI for anything important.
  2. Is this faster? If editing AI output takes longer, skip it.
  3. Does my voice matter? If authenticity matters, write it yourself.
  4. Would I disclose this? If you’d hide the AI use, reconsider.
  5. Am I learning? If you should learn this skill, don’t outsource it.

AI as a Crutch

Watch for patterns:

  • Using AI for everything
  • Feeling unable to work without AI
  • Losing skills you used to have
  • Avoiding difficult thinking

AI should enhance your capabilities, not replace them.

The Balance

AI is a powerful tool. So is a hammer.

You don’t use a hammer for everything.

Best use of AI: Amplifying your abilities, handling routine work, providing starting points.

Worst use of AI: Replacing thinking, avoiding effort, misrepresenting work.

Use AI deliberately. Not by default.

Related Articles