Use the BBC Ideas video How AI actually works with this guided discussion to help students understand what AI does in practice: pattern prediction, not human-style thinking.
AI Is Already Here!
BBC Ideas + guided discussion: what AI is actually doing when it gives confident answers.
- Lesson time
- 20 min
- Fits
- Assembly (20 min)
- Key stage
- KS3, KS4, KS5
- Format
- Presentation
- Downloads
- 21
- Views
- 220
Download PDF
21 downloads 220 views
Before you teach
Get ready
- Open the PDF (or the SlideForge deck) and check the BBC Ideas film plays with sound in the hall: youtube.com/watch?v=E4bvQZRC6Bo
- Have the video open in a browser tab, so you can switch to it on slide 5.
- Optional: prepare one live AI prompt and one fact-check example to show.
Teacher notes
Reinforce this line: "Useful prediction is not the same as human understanding." Encourage students to separate fluency from accuracy. Prompt examples where AI is useful (drafting, summarising) and where verification is non-negotiable (facts, citations, health/legal/finance). The five Being AI-Smart points on slide 5 work well as a printed takeaway card.
Learning objectives
- Understand that AI predicts patterns rather than ‘thinking’.
- Recognise the difference between pattern prediction and understanding.
- Learn why AI hallucinations occur and why verification is essential.
The lesson
-
Starts at 0:00
Have the title up as students come in.
-
Starts at 1:00
Hook: ask for a show of hands — has anyone used AI today? Then ask whether AI “understands” what it writes.
Teacher tip
Most hands stay down at first. The next slide changes that.
-
Starts at 2:00
Go through the four everyday uses, then ask again who has used AI today.
Teacher tip
Card C is the bridge to the film: a chatbot is autocomplete at enormous scale.
-
Starts at 3:00
Play the BBC Ideas film “How AI actually works”.
Teacher tip
Ask them to listen for one thing the AI does NOT do.
-
Starts at 8:00
Discuss: AI as pattern prediction — advanced autocomplete — against genuine understanding. Take two or three answers from the hall.
Teacher tip
Row 3 is the surprise: a model has no way of noticing that it does not know.
-
Starts at 12:00
Explain hallucinations: plausible wording can still be wrong, so verify important facts.
Teacher tip
If you prepared a fact-check example, show it here.
-
Starts at 14:00
Present the three Being AI-smart habits and invite responses from the hall.
-
Starts at 17:00
Exit check: ask students for one safe use of AI and one use that must be fact-checked.
-
Debate it (optional — now or next lesson): read the motion and take a first vote — for, against or not sure yet.
Teacher tip
Challenge card: What if the warnings appeared so often that everyone started ignoring them?
-
Split the class for and against. Each side builds its case from the three points, then hear two speakers a side and vote again.
Teacher tip
Ask who changed their mind, and what changed it.
-
Starts at 19:00
Signpost reporting and support resources, and leave them up as the hall empties.
Teacher tip
Name the actual person — tutor, head of year or safeguarding lead — rather than leaving it to the slide.
Big question
If AI can sound confident, what checks should we use before trusting its answer?
Adapting the lesson
Support
For pupils who need more help
Use sentence stems: 'AI predicts…', 'AI does not know…', 'I would verify by…'.
Stretch
For pupils ready to go further
Ask students to create one prompt likely to cause hallucination and redesign it to reduce risk.
SEND
For additional needs
Allow paired response cards instead of whole-class verbal answers.
In every lesson
Debate it
For the National AI Conversation
Turn it into a motion: reword the big question as "This house believes…". Split the class for and against, give sides five minutes to prepare, hear two speakers each, then take a free vote.
National AI Conversation
Set up the debate
The same topic, pitched for each age group. Pick your pupils' pathway: the motion, a prompt to get them going, and three points each way to start their preparation.
Primary · Years 5 and 6
The motion
Should you check everything AI tells you?
Sentence starter: I think you should / don't need to check because ___.
For
- AI can make things up and still sound very sure.
- Checking with a book or a trusted adult helps you find the truth.
- If you share something wrong, other people might believe it too.
Against
- Checking everything would take a very long time.
- For fun things, like a silly story, it does not matter if it is wrong.
- AI gets lots of simple things right.
- In class
- Pupils move to a corner of the room or vote with a show of hands, then tell a talk partner why.
- As a full debate
- Shorter speeches, with more time for questions from the floor and the final vote.
Secondary · Years 7 to 11
The motion
AI chatbots should have to warn you every time an answer might be made up.
Challenge card: What if the warnings appeared so often that everyone started ignoring them?
For
- Hallucinations happen regularly, and fluent answers make them hard to spot.
- A warning reminds people to verify before they trust or share.
- Companies already warn about other risks, such as age limits and gambling.
Against
- The system often cannot tell when it is wrong, so the warnings would be guesses too.
- Constant warnings cause warning fatigue, and people stop reading them.
- Learning to verify is a skill we need anyway, warning or not.
- In class
- Small groups with talking roles, Builder, Challenger and Summariser, then the challenge card tests their view. Shorter for Key Stage 3.
- As a full debate
- Three-minute opening cases, two-minute rebuttals, and questions from the floor.
Post-16 · Sixth form and college
The motion
This house would require AI tools in schools to guide students rather than give answers.
The tension: efficiency vs learning. Research: what does the evidence say about AI tutoring and learning outcomes?
For
- Learning comes from working things out; ready-made answers skip the thinking that builds understanding.
- Guiding tools reduce the risk of students copying a confident hallucination into their work.
- Schools have a duty to teach verification, and answer machines undermine it.
Against
- Students will use answer tools at home anyway, so schools should teach them to use them well.
- Sometimes a clear worked answer is the fastest way to learn, as with a model essay.
- A rule for every tool is hard to enforce and may hold back students who use AI responsibly.
- In class
- A formal debate, with points of information offered during the opening speeches.
- As a full debate
- Longer opening cases, and a show of hands or a digital vote at the end.
Key words
- Pattern prediction
Choosing likely next words from learned data patterns.
- Understanding
Human-style reasoning about meaning, context, and truth.
- Hallucination
A fluent but incorrect AI output produced from weak or ambiguous pattern matches.
- Verification
Checking outputs against reliable sources before trusting or sharing.
Take it further
- Next lesson Compare two AI answers to the same question and identify which one is better supported by evidence.
- Cross-curricular Create a class checklist for responsible everyday AI use.
Materials
- Video How AI actually works — BBC Ideas (YouTube)
- Slides AI Is Already Here! — Assembly slide deck