Free Orientation

Before You Write a Single Prompt

AI isn't magic or intelligence. Understanding what it actually is — and isn't — changes everything about how you use it. This free orientation gives you the foundation that most people skip.

No account required
15 minute read
Essential knowledge

You Write a Prompt

"Write me a marketing email"

AI Predicts Next Words

Pattern matching, not understanding

Output Looks Confident

But may be wrong — you must verify

What AI Actually Is

Modern AI systems like ChatGPT are pattern prediction engines. They've analyzed billions of documents and learned to predict what word comes next in a sequence. That's it. That's the whole trick.

When you ask AI a question, it doesn't "know" the answer — it predicts what a good answer would look like based on patterns in its training data. This is why AI can sound incredibly confident while being completely wrong.

Understanding this changes everything. You stop treating AI like an all-knowing oracle and start treating it like what it is: a powerful tool that requires human judgment.

Pattern Prediction

AI predicts the next word in a sequence, thousands of times per response.

Trained on Text

Billions of documents, websites, and books — patterns in human writing.

Sounds Confident

Trained on confident-sounding text, so it mimics that confidence.

No Real Understanding

No memory, no values, no judgment — just statistical patterns.

Why AI Fails

AI doesn't fail randomly — it fails in predictable, systematic ways. Once you understand these failure modes, you can anticipate them and protect yourself from making costly mistakes.

These aren't rare edge cases. Every AI user encounters these problems regularly. The difference between a competent AI user and an incompetent one is recognizing when they're happening.

Hallucination

Confidently generates false information — citations that don't exist, facts that are wrong.

Sycophancy

Agrees with you even when you're wrong, tells you what you want to hear.

Context Limits

Forgets earlier instructions in long conversations, loses track of what you asked.

Bias

Reflects prejudices and blind spots present in its training data.

Your Responsibility
as the Human

AI is a powerful tool, but tools don't have judgment. When you use AI in your work, you remain accountable for the results.

You Verify

AI doesn't know what's true. Every claim must be checked before you rely on it.

You Judge

AI optimizes for plausible, not good. Quality judgment is always human work.

You Decide

AI has no values or priorities. Every decision about what matters is yours.

You Account

AI takes no responsibility. When things go wrong, you own the outcome.

Quick Self-Assessment

Answer a few questions to get a personalized recommendation for where to start your learning journey.

I have used AI tools like ChatGPT, Claude, or similar:
I feel confident that AI outputs are accurate:
I understand why AI sometimes produces incorrect information:

Your Recommendation

Start with Micro-Labs

Ready to Build Real Skills?

Now that you understand what AI is and isn't, it's time to put that knowledge into practice. Start with our first Micro-Lab and learn to transform vague questions into precise instructions.