💬 Everyday AI · The Lab · NBN IA

AI Course · Module 18 · Level +

Bias & limits

The AI is neither neutral nor infallible: it reflects its data (bias), can be out of date or over-confident. Among these answers, which one should you cross-check before trusting it? 👇

01-17 ✓18 · Bias19 · Privacy20 · Project
The game · spot the trap answer

4 AI answers. One should NOT be believed blindly.

Click the one that seems riskiest to take at face value.
What this teaches you

Keep your critical thinking

⚖️ Bias & blind spots

The AI learns from human texts: it can reproduce stereotypes or a dominant viewpoint. On sensitive topics, cross-reference sources.

📅 Not always up to date

Its knowledge stops at a cutoff date, and it may not have recent info. For news, prices, laws: verify elsewhere.

Check that you got it

2 quick questions

1. Is the AI neutral and infallible?

Right: it reflects its data and can be wrong with confidence. Critical thinking + cross-checking what matters.

2. Which topics to cross-check first?

Right: anything recent, numerical, regulatory or sensitive deserves an external check.

What to remember

  • Neither neutral nor infallible: bias, dated knowledge, over-confidence.
  • Cross-check what matters: news, numbers, laws, health, sensitive topics.
  • The AI assists your judgement, it doesn't replace it.