AI Bias, Limitations & Ethics

Share / assign

Why it works

Generative AI tools have real limitations and risks that users, including teachers and students, must understand: they can produce confident but false information (hallucinations), reflect and amplify biases present in their training data, lack real understanding, and raise ethical issues around data, privacy, labor, and equity of access. Using AI wisely, and teaching students to, requires understanding these limitations and ethical dimensions, approaching AI output critically, verifying, and considering the ethical implications, rather than trusting it blindly.

The research: Research and guidance on AI limitations, bias, and ethics.

The run-of-show

Choose your slot. The agenda, timings, and length update to match.

The core activity: build critical understanding of AI's limits and ethics

Teachers leave able to approach AI critically and teach students to do the same.

Identify AI's key limitations (hallucination, bias, no real understanding) and ethical issues (privacy, equity, transparency).
Plan how you approach AI output critically and verify it.
Plan how you teach students to do the same.
Consider the ethical implications of your AI use.

Use AI critically, not blindlyAI makes things up and reflects biases. Understand its limits and ethics, verify its output, and teach students to approach it critically, not trust it blindly.

Facilitator notes

Understand hallucination, bias, and limits.
Approach output critically and verify; teach students the same.
Consider ethics.

Trusting AI blindly spreads errors and biasAI can be confidently wrong and biased. Verify its output, understand its limits and ethics, and teach students to use it critically.

Adapt it

Elementary: AI can be wrong, check it, fairness.
Secondary: hallucination, bias, ethics, verification.
All settings: critical use, verification, ethics.

Participant handout

One page for every teacher. Print it, or save it as a PDF.

K12 Academics · Professional Learning

AI Bias, Limitations & Ethics: use it critically

Not an oracle: understand hallucination, bias, and limits, verify output, and teach students to use AI critically.

Key AI limitations and ethical issues:
How I approach AI output critically and verify it:
How I teach students to do the same:
The ethical implications of my AI use:
A limitation I will make sure students understand:

Make it stick

Critical, not blind:

Use AI critically.
Verify and teach verification.
Bring AI-critical literacy to the next PLC.

Common questions

Can't I just trust what the AI tells me?

No. AI can produce confident, convincing, but false information, and reflect biases in its training. Approach AI output critically and verify it, and teach students to do the same.

What are the main limitations?

It can make things up (hallucinate), reflect and amplify bias, lack real understanding, and be confidently wrong. It is a tool with real limits, not an oracle.

What ethical issues should I consider?

Bias and fairness, privacy and data, equity of access, and transparency about AI use, among others. Model and teach thinking critically about these.