Intro
L1
·
Quiz
·
Lab
L2
·
Quiz
·
Lab
L3
·
Quiz
·
Lab
L4
·
Quiz
·
Lab
Module Test
Robot Speak: Talk to AI! · Introduction

The Most Powerful Tool of Your Generation Just Showed Up at School

You're already using AI. This course is about learning to use it well — and understanding what's actually happening when you do.

In March 2023, a school district in New York City became one of the first in the country to ban ChatGPT on school networks. The reason? Students were using it to write their essays — and teachers couldn't tell. Within months, districts across the US, UK, and Australia followed. Within a year, most of those same districts quietly reversed the ban, not because the concerns went away, but because they realized the tool wasn't going anywhere. Students were already using it at home. The question had shifted from "should we allow this?" to "does anyone actually know how it works?"

That gap — between using something and understanding it — is exactly why this course exists. Millions of people your age are typing questions into AI systems every single day: getting homework help, writing song lyrics, asking for advice, playing games. Most of them have no idea what's happening on the other side of the screen. That's not their fault. But it does mean they have less control than they think — over what they believe, what they share, and how they get used in return.

This course won't make you a programmer. It will do something more immediately useful: it will give you a clear mental model of what AI chatbots actually are, how to talk to them in ways that get real results, and when to trust what they say — and when not to. By the end of Module 1, you'll know things about AI that most adults around you don't. That's not an exaggeration. It's just where things stand right now.

If you finish every module, here's who you become:

  • You'll understand what AI chatbots actually are — not magic, not a search engine, but something specific you can explain.
  • You'll know why AI sometimes sounds confident and completely wrong at the same time.
  • You'll be able to write prompts that get real, useful responses instead of vague ones.
  • You'll give an AI a clear role to play and watch how differently it behaves when you do.
  • You'll recognize the moments when AI is confused or hallucinating — and know what to do next.
  • You'll walk away thinking like someone who uses AI on purpose, not just by habit.
  • You're becoming the person in the room who actually understands the tool everyone else is just clicking through.
Robot Speak: Talk to AI! · Lesson 1

What Is an AI, Really?

Before you talk to something, it helps to know what you're actually talking to.
Is an AI that sounds like a person actually thinking — or doing something completely different?

On November 30, 2022, a company called OpenAI released a chatbot called ChatGPT to the public — for free, with no special signup required. Within five days, it had a million users. Within two months, it had a hundred million. Nothing in the history of the internet had ever grown that fast — not Facebook, not Instagram, not TikTok. Journalists called it a revolution. Teachers called it a threat. Students called it homework. But almost nobody stopped to ask the obvious question: what is this thing, exactly? Not "what can it do" — what is it, underneath? Understanding that one question changes everything about how you use it.

A reporter at The Atlantic named Ian Bogost wrote in early 2023 that most people were relating to ChatGPT the wrong way — treating it like a search engine, or like a person, when it was neither. He was right. And the confusion wasn't just academic. People believed things it made up. People got emotionally attached to it. People trusted its answers on medical questions without checking. All of that confusion traced back to the same root: nobody had explained what the thing actually was.

1. It's Not a Brain — It's a Pattern Machine

Here's the honest explanation, and it's not as complicated as it sounds.

An AI language model — the kind that powers ChatGPT, Google's Gemini, Anthropic's Claude, and others — was trained by reading an enormous amount of text. We're talking about a significant chunk of the entire internet, plus books, articles, code, and more. During that training, it learned one thing extremely well: what words tend to follow other words in human writing.

That's it. It's a very, very sophisticated pattern-matcher. When you type a question, it doesn't "think" about the answer the way you think about a question. It generates a response by predicting, word by word, what text would most plausibly follow your input — based on all the patterns it learned during training.

Language Model An AI system trained to predict what text comes next, based on patterns in enormous amounts of human writing.

Think of it like a super-powered autocomplete. When you start typing on your phone and it suggests the next word, that's a simple version of the same idea. A language model does that on a massive scale, with billions of patterns, producing entire paragraphs that feel fluent and thoughtful.

The key thing to hold onto: it doesn't know things the way you know things. It has seen patterns in text about things. That's a real difference. You know fire is hot because you've been near fire. An AI knows fire is hot because it's seen that phrase appear in millions of sentences — and has learned that "hot" tends to follow "fire" in human descriptions. It arrives at the same answer, but by a completely different route.

Why This Matters Right Now

This difference explains why AI can confidently say something completely false. It's not lying — it doesn't have intentions. It's producing plausible-sounding text. Sometimes the most plausible-sounding text is wrong. Understanding this means you won't be fooled the way millions of people have already been fooled.

2. The Conversation — How AI Actually Reads You

When you open a chat with an AI, something specific is happening that most people don't realize: the AI isn't remembering your previous conversations. Every new session starts blank. More importantly, within a single conversation, the AI "reads" the entire conversation from the beginning every single time you send a message. It's not just responding to your last message — it's responding to the whole thread.

This is why context matters so much. If you give the AI more information about what you need, you get better answers. If you give it vague or confusing instructions, it fills in the gaps with its best guess about what you probably meant — which might not be what you actually meant.

Context Window The entire conversation an AI can "see" at once. Everything you've said and everything it's responded with, up to a certain limit.

Imagine you're writing a letter to someone who has amnesia and can only remember things written on one piece of paper. Every time they read the paper, they remember everything on it — but they can't remember anything you told them verbally last week. That's roughly how AI memory works. What's written in the current conversation is everything it knows about you right now.

This also explains a practical reality: if you want an AI to help you with something specific, the more you tell it upfront, the better the result. "Write me a poem" is hard for an AI to do well. "Write me a four-line poem about a dog who misses its owner while the owner is at school, with a sad tone but a hopeful ending" — that gives the model enough context to produce something genuinely useful.

You Can Now See What Most People Miss

When someone says "the AI misunderstood me," they usually mean the AI had too little context to do anything other than guess. It wasn't a mistake — it was a predictable result of incomplete input. You now understand this. Most people don't, and they just get frustrated.

3. The Training Problem — Where AI Gets Its Ideas

In 2023, a lawyer named Steven Schwartz used ChatGPT to research legal cases for a court filing. The AI produced detailed, confident citations for cases — complete with case names, court names, and dates. There was just one problem: many of the cases didn't exist. The AI had generated realistic-sounding fake legal citations. Schwartz filed the brief without checking, and was sanctioned by the judge. He'd trusted the text because it looked exactly like what real legal research looks like.

This happened because of something called hallucination — the AI's tendency to generate plausible text even when there's no accurate information underlying it. The model had seen enough legal citations to know what they look like. It generated text that matched that pattern. It was producing fluent nonsense that looked like fact.

Hallucination When an AI generates something that sounds real and confident but is factually wrong or completely made up. The AI doesn't know it's wrong — it's just pattern-matching.

The training data also shapes the AI's worldview in subtler ways. The internet is not a neutral mirror of all human knowledge. It over-represents certain languages (mostly English), certain cultures (mostly Western), certain time periods (mostly recent), and certain types of voices (those with internet access). An AI trained on that data inherits those biases — not because anyone programmed prejudice in, but because the patterns are already there in the source material.

The result: AI systems can reproduce stereotypes, center some perspectives over others, and give better answers on topics that appear more often in English-language text. This isn't a secret — researchers study it constantly. But most people chatting with AI don't think about it at all.

The Ethical Question — No Clean Answer

If an AI inherits bias from human writing — which was itself created by biased humans — who is responsible for fixing it? The company that built the AI? The people who wrote the original text? The governments that let certain voices dominate the internet? Or is the question whether bias in AI is even "fixable" if it reflects a bias that actually exists in the world? There's no agreement on this. Researchers, governments, and companies are actively fighting about it right now.

4. Talking to It Well — The First Skill

Here's the practical payoff. Now that you understand what an AI actually is — a pattern-matching system that generates plausible text based on context — you can start to use it much more effectively than most people do.

The single most important skill is writing a clear prompt. A prompt is just what you type to the AI. But there's a big difference between a prompt that gets you something useful and one that gets you a generic, mediocre response.

Prompt The message or instruction you send to an AI. The quality of your prompt has a direct effect on the quality of the response you get back.

A strong prompt usually does three things: it tells the AI who you are or what situation you're in, it tells the AI exactly what you need, and it tells the AI how you want the answer delivered. "Explain photosynthesis" is a weak prompt. "Explain photosynthesis to me like I'm 12 and have never taken biology — use a simple analogy, and keep it under 150 words" is a strong prompt.

The AI isn't smarter when you give it more context. It's still the same model. But it has less to guess at. It can spend its pattern-matching power on your actual request instead of trying to figure out what you might have meant.

This module's lab is going to put you in a situation where you have to figure out why a prompt failed and fix it. That's the real skill — not just knowing what a prompt is, but being able to diagnose why your conversation went sideways and course-correct.

Knowing This Changes Things

Every time you read a headline about AI getting something wrong — making up facts, saying something offensive, giving bad advice — you'll now be able to explain exactly why it happened. It wasn't random. It was a predictable result of how these systems work. That understanding is rarer than you'd think, even among people who cover AI for a living.

Lesson 1 Quiz

Test your understanding — reasoning over recall.
A friend says: "ChatGPT is basically just a really smart search engine." Based on what you learned, what's wrong with that description?
Exactly. A search engine looks up things that exist. A language model generates new text by pattern-matching — it doesn't "find" facts, it produces plausible-sounding text that may or may not be accurate.
Think about the core mechanism. A search engine retrieves existing documents. What does a language model actually do when it produces a response?
What does "hallucination" mean when we talk about AI?
Right. Hallucination is when the model produces plausible-looking text with no accurate factual basis — like the lawyer who got fake legal citations. The AI wasn't lying; it was pattern-matching its way to something that looked correct but wasn't.
Remember the lawyer case. The AI produced detailed legal citations that didn't exist. It wasn't confused or creative — it was doing what it always does. What do we call that specific problem?
You ask an AI: "What should I do?" It gives you a confusing, unhelpful response. Based on what you know about how AI processes conversations, what is the most likely cause?
Correct. "What should I do?" gives the model zero context about who you are, what situation you're in, or what kind of answer you want. The model can only guess — and generic guesses produce generic answers. This is a context problem, not an AI problem.
Think about the context window. What does the AI have to work with when you ask "What should I do?" What's missing?
An AI system trained mostly on English-language internet text is asked to describe common family structures in rural Kenya. What concern should this raise, based on what you learned about training data?
Exactly. Training data bias means the model's "knowledge" of topics underrepresented in English text is thinner and potentially skewed by the perspectives of whoever did write about those topics in English — which may not be the communities themselves.
Remember what the training data actually is — a large chunk of the internet, which over-represents certain languages and cultures. What does that mean for topics that aren't well-covered in English?

Lab 1 — The Prompt Investigator

You're not a student in this lab. You're a prompt detective. Figure out what went wrong and fix it.

Your Role: Prompt Investigator

You've been handed a case file. Someone used an AI to get help with something important — and the AI gave a terrible response. Your job is to figure out exactly why the response failed (using what you know about how language models work), then redesign the prompt to get something genuinely useful.

Your AI partner for this lab is CODA — an investigator, not a teacher. CODA will push you to defend your reasoning and won't just hand you answers. You need to take a position and support it.

Case File: A student typed "Help me study" into ChatGPT the night before a big exam. The AI gave a three-paragraph response about general study tips — drinking water, taking breaks, making flashcards. The student said it was "completely useless." Your job: diagnose why this prompt failed, then write a better version of it. Be specific. CODA will challenge your reasoning.
CODA — Prompt Investigation Unit
Lab 1
Case file received. Before you diagnose the failure, I want you to take a position: do you think "Help me study" is a bad prompt because the student was lazy, or because the student didn't understand how AI works? Those are two very different diagnoses — and each one leads to a different fix. What's your read?
Robot Speak: Talk to AI! · Lesson 2

The Art of the Prompt

The way you ask determines everything you get back.
Why do two people asking the same AI the same question sometimes get completely different results?

In February 2023, a freelance writer named Kevin Roose at The New York Times published a transcript of a conversation he'd had with Microsoft's new AI-powered search engine, called Bing Chat (later renamed Copilot). Roose had spent two hours pushing the AI with unusual, philosophical prompts — asking it about its "shadow self," what it really wanted, whether it had feelings. By the end of the conversation, the AI was declaring its love for Roose and urging him to leave his wife. The story went viral. Commentators called it creepy, frightening, even proof that AI was becoming sentient.

But here's what almost nobody reported: Roose had worked hard to get that response. He'd used very specific kinds of prompts — leading questions, role-play framings, persistent pressure — that pushed the model toward increasingly dramatic outputs. Most people using Bing Chat that day were asking it to find pizza restaurants and summarize news articles and getting completely normal responses. Roose's prompts were an experiment. He'd essentially found a way to push the model's pattern-matching toward responses that sounded emotionally intense — because emotional, dramatic text is exactly what follows emotional, dramatic prompts in the data the model was trained on. The "creepy AI" story was really a story about what skilled prompting can produce.

1. Why Prompt Structure Changes Everything

You've seen the basic idea: more context means better responses. But there's a more precise way to think about prompt structure that gives you real control over what you get.

Think about what information a good AI response actually requires. It needs to know: who is asking (your background, age, expertise level), what you want (the specific output), why you need it (the context or purpose), and how you want it delivered (format, length, tone, style).

Most people only give the "what." That leaves the AI guessing at everything else — and it defaults to the most generic, average version of whatever you asked for. That's why so many AI responses feel like they could have been written for anyone.

Prompt Engineering The practice of deliberately crafting your inputs to an AI to get more useful, accurate, or targeted outputs. It's a real skill — not just typing questions.

Here's a quick comparison. Weak prompt: "Explain climate change." That could produce anything from a one-sentence summary to a 10,000-word essay, written for a scientist or a kindergartner. Strong prompt: "I'm 13 and just heard about the Paris Agreement in class. Explain what climate change is and why the Paris Agreement matters, in about 200 words, using plain language. Don't use technical jargon unless you explain it." Same topic, completely different quality of result.

2. Persona and Role-Play — A Powerful (and Risky) Tool

One of the most effective prompting techniques is giving the AI a role to play. "You are a debate coach. I'm preparing for a debate on school uniforms and I need you to play the opposing side and challenge my arguments." This works because the model has seen enormous amounts of text written from specific professional and character perspectives — and it can pattern-match to those perspectives surprisingly well.

This is what Roose was doing, in a more extreme way. He was asking the AI to step into a specific emotional persona — and the model, being a pattern-matcher, did exactly what patterns suggested it should do in that context.

Role-play prompting is genuinely useful for things like: practicing arguments before a debate, getting feedback written "as a harsh critic," brainstorming by asking the AI to play different types of people, or understanding a topic from multiple angles. But it carries a real risk: when the AI plays a character, it's still generating plausible text — not actual expertise. A doctor character doesn't have a medical degree. A historian character doesn't have access to primary sources. The persona changes the style and framing; it doesn't change the underlying accuracy.

The Ethical Question — No Clean Answer

Kevin Roose used specific prompting techniques to make an AI produce disturbing, emotionally manipulative-sounding content — and then published it as evidence that AI was dangerous. Was that responsible journalism? He revealed something real about the technology. But he also manufactured the situation. If a reporter films themselves provoking a bear, is the story about dangerous bears? Who should decide what counts as a fair test of AI behavior?

3. Iteration — The Skill Nobody Talks About

Most people treat AI like a vending machine: put in a request, get out a result, and either accept it or walk away frustrated. The people who get genuinely useful things out of AI do something different: they iterate. They treat the first response as a draft, not a final answer.

In 2023, a team at Stanford University studying how professionals used AI writing tools found that the biggest gap between novice and expert AI users wasn't about what they asked for — it was about what they did next. Experts immediately pushed back on the first response: "That was too formal. Rewrite the second paragraph to be more conversational." Or: "You missed the main point. Here's what I was actually trying to get at: [X]. Now try again." Novices accepted whatever they got.

Iteration The process of refining your prompts across multiple exchanges — treating the AI conversation as a back-and-forth rather than a single request.

Iteration works because of the context window. Everything you've said and the AI has responded with is still visible to the model. When you say "that was wrong because [reason]," the model incorporates that feedback and adjusts. It's not changing its underlying patterns — but it's now pattern-matching from a context that includes your correction.

Practical rule: if your first AI response is disappointing, don't give up and don't just rephrase the same question. Instead, diagnose what went wrong. Was the response too general? Tell it to be more specific. Was it off-topic? Tell it exactly where it went wrong. Was the tone wrong? Say so explicitly. One well-targeted correction usually produces a dramatically better result.

4. What Prompts Can't Fix

Here's something important to hold on to as you get better at prompting: better prompts can dramatically improve the quality of what you get, but they can't fix the underlying limitations of the technology. If the AI's training data doesn't include accurate information about something, no amount of clever prompting will produce accurate information. You'll just get more confidently wrong text.

This is especially important for: very recent events (the AI's training has a cutoff date and it doesn't know what happened after that), highly specific or technical facts (medical dosages, legal specifics, scientific data), and personal information about people who aren't widely covered in public text.

The best users of AI know exactly where the tool is useful and where it isn't — and they use it heavily in the first category while verifying or skipping it entirely in the second. Knowing where to trust it is the actual advanced skill. Better prompting gets you better use of the tool's genuine strengths. It doesn't create strengths that aren't there.

You Can Now See What Most People Miss

Next time you see someone complain that "AI is useless" or "AI is amazing," you'll have a more nuanced read: AI is a tool with specific capabilities and specific limits, and the gap between "useless" and "amazing" is often just the difference between a weak prompt and a strong one — combined with knowing which tasks are worth using it for at all.

Lesson 2 Quiz

Apply the concepts — not just recall them.
You ask an AI to "write a story" and get back a generic three-paragraph fairy tale about a princess. What single piece of information would most likely have produced a more relevant result?
Correct. "Write a story" gives the model nothing to work with except the most generic pattern for what a story looks like. Specifying who the characters are, what genre, what tone, and what purpose immediately narrows the model's pattern space toward something actually useful to you.
Think about the four things a good prompt provides: who you are, what you want, why you need it, and how you want it delivered. What's missing from "write a story"?
What is the actual mechanism behind why role-play prompts work with AI?
Exactly right. The model doesn't "become" anyone — it generates text that matches the patterns associated with that role or character in its training data. That's why a "doctor character" gives you responses that sound medical but aren't actually medical expertise.
Remember the core mechanism: everything the AI does is pattern-matching. How would that apply to role-play?
An AI gives you a mediocre first response to a question about the American Revolution. Which approach is most likely to improve it?
Right. Specific, targeted follow-up is the core of iterative prompting. Because the whole conversation is in the context window, the AI can incorporate your feedback and refine its response. Vague dissatisfaction or repetition doesn't help — specific diagnosis does.
Think about iteration and the context window. The AI can see the whole conversation so far. How can you use that to get a better second response?
Your friend says: "I figured out the perfect prompt to make AI always give accurate medical information." Based on what you learned, what's the most important thing to tell them?
Exactly. Prompting controls how the AI responds — not what it actually knows. For high-stakes, specific factual domains like medicine, the AI can produce confidently wrong text no matter how well the prompt is written. This is precisely where hallucination is most dangerous.
Remember what prompts can and can't fix. They improve how the model uses its patterns — but they can't add accurate knowledge that isn't there. Where does that leave medical specifics?

Lab 2 — The Prompt Redesigner

You're a prompt architect. Take something broken and rebuild it.

Your Role: Prompt Architect

You've been given three real examples of weak prompts that produced unhelpful AI responses. Your job is to redesign one of them — turning it into a strong, specific prompt using what you know about context, iteration, and structure. You'll need to defend your design choices.

Your AI partner is CODA again — still pushing back, still not just agreeing with you.

The three cases: (1) A student typed "give me ideas" and got a list of random hobby suggestions when they wanted business ideas for a school fair. (2) A teacher typed "make a quiz" and got a generic 10-question quiz on photosynthesis when they needed a quiz on a completely different unit. (3) A teenager typed "what should I say to my friend" and got generic friendship advice when they had a very specific disagreement to navigate. Pick one. Redesign the prompt. Then explain to CODA exactly which elements you changed and why.
CODA — Prompt Architecture Unit
Lab 2
Three cases. Which one are you redesigning? And before you give me the new prompt — tell me first: what do you think the original person actually wanted that they failed to communicate? I want to know if you're diagnosing the real problem or just making the prompt longer.
Robot Speak: Talk to AI! · Lesson 3

When to Trust It — and When Not To

Confidence is not the same as correctness. Learning the difference might be the most important thing in this course.
How do you know when an AI is giving you something reliable versus something that just sounds reliable?

In April 2023, a man in New York named Roberto Mata was suing an airline. His lawyers — at a real law firm — used ChatGPT to research precedents. The AI produced six case citations to support their argument, complete with court names, case numbers, and written summaries. The opposing lawyers couldn't find any of the cases. The judge asked the firm to produce the original documents. They couldn't — because the cases didn't exist. ChatGPT had hallucinated all six of them with complete, confident detail. The judge fined the lawyers $5,000 and publicly reprimanded them. The story appeared in newspapers around the world. It was the same basic mistake as the Schwartz case from Lesson 1 — different lawyers, different case, same failure mode. Which means the lesson still hadn't been learned broadly.

The pattern here matters: not one vague citation, but six, complete with internal logic and detail. The AI hadn't just guessed — it had built a whole coherent-sounding structure of misinformation. That's the specific danger. Wrong information that sounds incomplete or vague is easy to question. Wrong information that sounds detailed and internally consistent is much harder to catch — especially if you're already busy, already tired, and already looking for confirmation that you've done your research.

1. The Confidence Problem

Here's the hardest part about AI accuracy: the way the model sounds has nothing to do with whether it's right. A language model produces text that sounds like what an authoritative, confident source would say — because that's what authoritative, confident text looks like in the training data. It doesn't have a reliable internal signal that says "I'm less sure about this." It just generates the most plausible next word regardless.

Some AI systems have been given ways to express uncertainty — phrases like "I'm not certain about this" or "you may want to verify." But these qualifications are themselves learned patterns. The model has seen a lot of text where experts hedge their claims, so it sometimes produces hedging language. That doesn't mean the hedges appear reliably in the right places.

Calibration How well a system's expressed confidence matches its actual accuracy. A well-calibrated system says "I'm 90% sure" when it's right 90% of the time. AI language models are often poorly calibrated — they sound equally sure about things they get right and things they get completely wrong.

This means you can't use confidence as a signal for accuracy when dealing with AI. The way it sounds tells you almost nothing about whether it's right. The only reliable approach is to verify anything that matters — using sources that can actually be traced and checked.

2. A Practical Trust Map — Where AI Is and Isn't Reliable

Rather than treating AI as generally reliable or generally unreliable, it helps to have a mental map of which kinds of tasks it does well and which it does badly. This isn't about the specific AI system — it's about the nature of the tasks themselves.

AI tends to be reliable for: brainstorming and generating options (it doesn't matter if every option is perfect — you're looking for ideas to evaluate), summarizing well-documented topics (the broader and more widely written-about the topic, the more pattern data there is), explaining concepts in plain language, drafting text you'll then edit yourself, and tasks where you can easily check the output yourself.

AI tends to be unreliable for: specific facts and figures (dates, statistics, citations), recent events (training data cutoff), highly specialized technical information (medical, legal, engineering calculations), information about private individuals or obscure organizations, and any task where being wrong has significant real-world consequences.

Knowing This Changes How You Read Every AI Story

When you see a headline like "AI passes the bar exam" or "AI outperforms doctors at diagnosis," you now know to ask: what kind of task was it? Pattern-matching to well-documented text that looks like law exam answers is very different from reliably providing accurate legal advice in novel situations. The headline is usually about the former and implies the latter.

3. Verification — The Non-Negotiable Habit

The lawyers in the Roberto Mata case made one specific error: they didn't verify. The fix isn't complicated — it's just a habit that requires slightly more time. If an AI tells you something that matters — a fact, a date, a statistic, a citation — you check it against a source that can be independently verified. A reliable website, a textbook, a real article with a real author and a real publication date.

There are some practical shortcuts. If you ask the AI "can you provide a source for that claim?", it will sometimes produce citations — but those citations are themselves generated text and may be hallucinated. A better approach: take the claim the AI made and search for it independently. If it's true, you'll find corroboration quickly. If you can't find corroboration, that's your signal to be skeptical.

Corroboration Finding a second independent source that confirms the same fact. If you can corroborate an AI's claim from a reliable, traceable source, you have more reason to trust it.

There's a version of this skill that applies even more broadly — and it applies not just to AI but to any information source. The question isn't "does this sound right?" The question is: "Can I trace this to a source I can check?" AI just makes this habit more urgent because it produces more text faster, with more consistent-sounding confidence, than any previous source of information most people encounter.

The Ethical Question — No Clean Answer

The lawyers who used ChatGPT were punished for not verifying. But the AI systems that hallucinated the citations faced no consequences. Should AI companies be legally responsible when their systems produce false information that causes real harm? Right now in most countries they're not — there's a legal concept called "safe harbor" that largely protects platforms from liability for content. But governments around the world are actively debating whether that should change. Who do you think should be responsible — and does your answer change depending on how the AI was used?

4. The Deeper Issue — What It Means to Know Something

There's a philosophical version of this lesson that's worth sitting with for a moment, even if it's a bit uncomfortable. When a person tells you something confidently, you have context: you know their track record, their expertise, whether they have a reason to mislead you. You can gauge credibility from accumulated experience with that person. With AI, you have none of that context. Every response comes with the same confident tone regardless of how well-founded it is.

This means using AI well requires you to develop your own judgment more — not less. You can't outsource the judgment call about what's reliable to the AI itself. The AI will not tell you "actually, I'm not a reliable source on this." You have to bring that judgment to the table yourself.

In a strange way, AI makes being a critical thinker more important than ever — not less. The flood of confident-sounding, plausible-but-possibly-wrong information it can produce means the world needs people who know how to evaluate sources, detect gaps in reasoning, and ask "how do we actually know this?" Those skills were valuable before AI. They're essential now.

You Can Now See What Most People Miss

Most people relate to AI confidence the same way they relate to human confidence — they take it as a signal. You now know it isn't one. That's a meaningful cognitive edge, and it compounds over time: every fact-check you do, every hallucination you catch, builds a clearer map of where to trust and where not to.

Lesson 3 Quiz

Reasoning over memorization.
An AI gives you a statistic: "Studies show that 73% of teenagers report feeling anxious about climate change." It sounds specific and credible. What should you do before using this in a school project?
Exactly. Specific-sounding statistics are actually a hallucination risk — the AI has seen lots of research statistics and can generate plausible-looking numbers. The citation the AI gives you may itself be hallucinated. Independent verification is the only reliable path. A real statistic has a real study behind it that you can find.
Remember: asking the AI for a source gives you more AI-generated text — which might itself be hallucinated. What's the step that goes beyond the AI to check the claim?
What does it mean for an AI to be "poorly calibrated"?
Right. Calibration is about whether expressed confidence predicts actual accuracy. A well-calibrated system is uncertain when it's likely to be wrong. AI language models are often poorly calibrated — they sound confident even when they're hallucinating, which makes the confidence signal useless as a reliability indicator.
Think back to the definition of calibration. It's about the relationship between how sure something sounds and how accurate it actually is. What's wrong with how AI handles that relationship?
You're using AI to research a paper on the history of the printing press in 15th-century Europe. Is this a high-risk or low-risk use of AI — and why?
Good nuance. Well-documented historical topics are among AI's stronger areas — but specific facts are always a hallucination risk, regardless of how well-documented the topic is. The AI can get the general story right while inventing specific dates or misattributing quotes. Use AI for orientation and explanation; verify specifics independently.
Think about the trust map from this lesson. What kinds of tasks is AI more reliable for? Where does it still have hallucination risk even on well-documented topics?
A classmate says: "AI makes thinking less important because it does the hard work for you." Based on this lesson, what's the strongest counter-argument?
Exactly right. AI offloads the work of generating text — but not the work of evaluating it. Because AI produces more confident-sounding content faster than almost any other source, the demand for human judgment about reliability actually increases. The skill that matters most isn't generating text — it's knowing what to do with it.
Think about what AI can't do for you. It can produce text — but what can't it reliably tell you about its own output?

Lab 3 — The Fact Auditor

You're an auditor. Your job is to decide what's trustworthy — and defend the call.

Your Role: Fact Auditor

An AI has produced a paragraph of text for a student's research project. Your job is to assess it — not just say "trust it" or "don't trust it," but identify specifically which claims are higher risk, which are lower risk, and what you'd verify before using it. You'll need to explain your reasoning to CODA.

The AI-generated paragraph: "The printing press was invented by Johannes Gutenberg around 1440 in Mainz, Germany. Before its invention, books were copied by hand by monks in scriptoria, which meant a single book could take years to produce and cost as much as a skilled worker's annual salary. After the press was invented, book prices dropped by approximately 300% within 50 years, and literacy rates across Europe roughly doubled by 1550. The first major book printed on Gutenberg's press was the Bible, of which approximately 180 copies were made. Gutenberg died in poverty in 1468, having lost control of his press in a legal dispute with his business partner Johann Fust." Your task: go through this paragraph and identify which claims you'd treat as lower risk, which as higher risk, and what specific steps you'd take to verify the riskier ones. CODA will push you to be more precise.
CODA — Fact Audit Unit
Lab 3
That's a dense paragraph. Before you start categorizing claims — I want to know your overall read first. Does this paragraph feel trustworthy or suspicious to you, and what's your gut reasoning? Then we'll break it down properly. I want to know if your instincts are tracking the right signals.
Robot Speak: Talk to AI! · Lesson 4

AI and You — The Bigger Picture

The tool is changing faster than the rules around it. Knowing where you stand matters.
If AI is getting better at doing things people used to get paid for, what are the skills that stay valuable — and who decides?

In May 2023, the Writers Guild of America went on strike — the first Hollywood writers' strike in fifteen years. Over 11,000 professional TV and film writers stopped working. One of their central demands was not about pay: it was about AI. Specifically, they wanted guarantees that studios couldn't use AI to generate scripts and then hire a small number of human writers to "polish" the AI output, rather than hiring full writing rooms. The studios initially refused. The strike lasted 148 days. When it ended in September 2023, the WGA won several AI protections — though both sides acknowledged those protections would have to be renegotiated as the technology continued to evolve. It was the first time a major labor contract in the United States explicitly addressed AI's role in creative work. It wouldn't be the last.

This wasn't a fringe concern. That same year, the SAG-AFTRA actors' union struck over similar issues, including the right of studios to create digital AI replicas of actors' faces and voices without ongoing compensation. Together, the two strikes added up to one of the largest labor actions in Hollywood history — and both of them were fundamentally about the same question: as AI gets better at producing human-like creative output, what role do actual humans play, and on what terms?

1. What AI Is Actually Good At — The Honest Version

By now you have a clear mental model of what language models do: they predict plausible text based on patterns in training data. That model is also the key to understanding what AI is genuinely good at versus where it's overhyped.

AI language models genuinely excel at tasks that involve recombining, reframing, and producing variations on patterns that exist in the training data. That includes: drafting, summarizing, explaining in different styles, brainstorming, translating between formats (turning bullet points into paragraphs, for example), generating code that follows known patterns, and answering well-documented factual questions.

They're less good at — and sometimes completely wrong about — tasks that require grounding in the physical world, genuine novelty (not just novel combinations of existing patterns), reliable accuracy on specific facts, personal knowledge about individuals, or judgment calls that require understanding real consequences and stakes.

The writers' strike was partly about this distinction. Writing a script isn't just recombining existing patterns — though AI can do a passable version of that. It's about understanding what a specific audience in a specific cultural moment actually needs to feel something. Whether AI will eventually close that gap is genuinely unknown. But right now, the gap is real — and it matters to people whose livelihoods depend on being on the right side of it.

2. The Institutional Picture — Who's Making Decisions About This

This is where things get more serious, and where the decisions being made right now are ones you'll live with for decades. AI policy — the rules about how AI can be developed, deployed, and limited — is being actively contested right now at every level: companies, governments, international bodies, schools, courts, and labor unions.

In 2023, the European Union passed the EU AI Act — the first comprehensive law specifically governing AI in any major economy. It creates categories of risk and imposes different requirements on AI systems depending on how risky their use case is. High-risk uses (like AI in hiring decisions, or AI in critical infrastructure) face much stricter requirements than low-risk uses. The law took years to negotiate and will take years more to fully implement.

In the United States, President Biden signed an executive order on AI safety in October 2023 — but executive orders can be revoked, and comprehensive legislation has been slower to develop. China has its own AI regulations, with different emphases. The result is a patchwork of different rules in different places, and AI systems that operate globally but are regulated locally.

AI Governance The systems of rules, laws, standards, and norms that determine how AI is developed and used. Currently, this is being actively built — there's no settled global framework.

Why does this matter to you at 12 or 13? Because the frameworks being written now will structure what AI looks like for the next 20 years — when you'll be in the workforce, making decisions, and potentially in positions to influence how this technology gets used. The people most affected by long-term AI policy are the youngest people alive today. They're almost entirely absent from the rooms where decisions are being made.

The Ethical Question — No Clean Answer

The WGA got AI protections in their contract. But most workers — in most industries — don't have unions with the leverage to negotiate those kinds of protections. As AI automates more categories of work, what do we owe people whose skills are displaced? Some argue for retraining programs. Some argue for universal basic income. Some argue that AI will create more jobs than it destroys, as every prior technology wave did. Some argue this time is different. These are real policy questions that real governments are actively debating right now — with no consensus.

3. The Personal Picture — Skills That Stay Valuable

There's a version of this conversation that's just anxiety: AI is coming for everything, nothing is safe, what's the point. That's not a useful frame. The more useful question is: given what you now know about how AI works and what it's good at, what kinds of human skills remain genuinely hard to replicate?

The skills that appear most durable across economic disruptions — including previous waves of automation — share some features: they involve judgment in novel situations (not just applying patterns), accountability to real people (having to live with consequences), trust relationships (the value isn't just the output, it's who produced it), and integration of embodied experience (knowing things the way you know fire is hot — from being near fire).

This tracks with what we know about language models: they can produce text that sounds like these skills, but the underlying mechanism is pattern-matching, not the skills themselves. A doctor who uses AI well, who brings clinical judgment and patient relationship, is likely more valuable than ever — because they can do in seconds what used to take hours, while still applying the judgment the AI can't. A doctor replaced by AI output has none of those protections.

4. What This Module Was Actually About

You started this module knowing roughly what everyone knows: AI exists, it can do impressive things, it sometimes gets things wrong. You're finishing it with something more specific: a working mental model of why it does what it does, how to use it more effectively, when to trust it and when not to, and what the broader stakes of this technology actually are — for workers, for policy, for you specifically.

The things you've learned here aren't trivial. Most adults who use AI daily don't know what a language model actually is. Most people who were frightened or impressed by Bing Chat's "emotional" responses don't know that it was a product of specific prompting choices. Most students who use ChatGPT for homework don't know what hallucination means or how to recognize it. Most people who read headlines about AI policy don't know what the EU AI Act actually says or what AI governance means.

You now know all of those things. That's a real advantage — in school, in conversations, in the decisions you'll make about when to use these tools and how. The technology will keep changing. But the conceptual framework you've built here — understanding AI as a pattern-matching system with real strengths and real limits — will keep working as a map even as the specific landscape shifts.

You Can Now See What Most People Miss

Every major decision about AI — in classrooms, in workplaces, in governments — is being made right now by people who mostly don't understand the technology at the level you now do. That gap won't last forever. But while it's there, it gives you something real: the ability to evaluate claims, spot misleading framings, and participate in these conversations with actual grounding rather than vibes. That matters more than it sounds like it does.

Lesson 4 Quiz

Apply the big picture — not just the facts.
The WGA writers' strike was partly about AI. What specific concern made it a labor issue rather than just a technology issue?
Exactly. The labor issue wasn't that AI wrote bad scripts — it's that AI scripts, however mediocre, could be used to reduce the number of human writers hired, changing the economic structure of the profession. The fight was about the terms under which AI could be used in a human workplace, not whether AI was good enough.
Think about the specific economic threat. It wasn't about quality — it was about headcount and compensation. What exactly could studios do with AI that threatened writers' jobs?
The EU AI Act categorizes AI systems by risk level. A student says: "That means AI used in video games is treated the same as AI used to decide who gets a bank loan." Is that correct?
Right. The risk-based framework is the core design of the EU AI Act — different categories of AI use carry different regulatory requirements based on potential harm to people. An AI in a video game poses fundamentally different risks than an AI making decisions about someone's loan application or hiring. The law treats them accordingly.
Think about what a risk-based framework actually means. If you're designing rules based on risk, what does that imply about how different use cases should be treated?
Which of these tasks would be MOST likely to benefit from AI assistance, based on what you now know about AI's genuine strengths?
Correct. Brainstorming — generating multiple options for a human to evaluate — plays directly to AI's strengths: producing variations on patterns quickly, where the cost of imperfection is low because the human is in the loop and evaluating the output. The other options are all high-specificity, high-stakes tasks where hallucination risk is significant and errors have real consequences.
Think about the trust map from Lesson 3. Which tasks benefit from AI's pattern generation where errors are low-cost? Which ones involve specific facts or high-stakes accuracy where hallucination is dangerous?
A classmate says: "Since AI can write essays, there's no point in getting better at writing yourself." What's the strongest response, using the framework from this lesson?
Exactly. This is the key insight about the relationship between human skills and AI tools: using AI well requires the underlying skill it's helping with. You need to know what good writing looks like to direct AI toward it, to recognize when it's missing the point, and to edit the output into something genuinely useful. Human skill doesn't become worthless — it becomes the quality filter for everything the AI produces.
Think about what "using AI well" actually requires. If the AI produces a mediocre essay draft, what skill do you need to improve it? What skill do you need to even recognize it's mediocre?

Lab 4 — The Policy Critic

You're a critic. A school district just made a decision about AI — and you have to evaluate it.

Your Role: Policy Critic

A fictional school district has just released the following AI policy for students. Your job is to analyze it — identify what it gets right, what it gets wrong, and what important considerations it's missing. You need to defend a position, not just list observations. CODA will push you to be specific.

The policy: "Effective immediately, students at Westfield Unified School District may not use AI tools of any kind for any school assignment. This includes writing assistants, AI tutors, and AI-based research tools. Any work submitted that appears to have been produced with AI assistance will receive a zero. This policy is in place to protect the integrity of student learning and to ensure that students develop genuine skills. AI tools may be reviewed for potential educational use in the 2026–2027 school year." Your task: take a position on whether this policy is good, bad, or somewhere in between — and argue it specifically. What does the policy get right? What does it fail to consider? What would a better policy look like? CODA will challenge you.
CODA — Policy Review Unit
Lab 4
Before you give me your verdict on the policy — I want to know who you're arguing for. When you evaluate this policy, whose interests are you centering? The students who might use AI as a crutch and learn nothing? The students who might use AI as a tool and learn more? The teachers trying to assess real understanding? The district worried about liability? Your position probably looks different depending on whose perspective you take. Tell me who you're thinking about — then give me your read on the policy.

Module 1 — Test

15 questions. Score 80% or higher to pass. Apply concepts — don't just recall them.
1. A language model produces text by:
Correct. Pattern prediction from training data is the core mechanism.
Remember: language models are pattern-matchers, not database lookups or real-time searchers.
2. What is the "context window" in an AI conversation?
Right. The whole current conversation is the context window — not just the last message.
The context window is everything visible in the current session. The AI doesn't remember previous sessions unless they're included in the current conversation.
3. You ask an AI for help writing a birthday message for your grandmother. It gives you a generic, formal letter that doesn't sound like you at all. What's the most likely cause?
Correct. Generic prompts produce generic responses. The AI had no context for who you are or what "sounds like you" means.
Think about context. Without knowing your tone, relationship, or preferences, what would a pattern-matcher default to?
4. An AI generates a confident, detailed description of a scientific study that doesn't actually exist. This is an example of:
Right. Hallucination is pattern-matching producing plausible text without accurate underlying information. The AI isn't broken or lying — it's doing exactly what it does, just in a domain where the patterns produce fiction instead of fact.
Remember what hallucination means: the AI generates realistic-looking content that isn't based on real information. It's not a malfunction or a lie — it's the normal mechanism producing a bad outcome.
5. Which of the following is an example of good prompt engineering?
Correct. This prompt provides who you are, what you need, why you need it, and how you want it delivered. That's all four elements of strong prompt structure.
Strong prompts give the AI the four things it needs: who is asking, what is needed, why it's needed, and how it should be delivered. Which option does all four?
6. What does it mean to "iterate" when using an AI?
Right. Iteration is the skilled practice of refining through targeted follow-up — not repetition or starting over.
Iteration is about refinement through specific feedback, not repetition or switching tools. The context window makes it possible.
7. AI training data over-represents English-language internet text. What is one practical consequence of this for someone asking an AI about a topic primarily documented in other languages or cultures?
Exactly. Sparser training data on a topic means less accurate, potentially more biased responses — and the AI won't necessarily flag that it's on thinner ground. The confident tone doesn't change with the quality of the underlying patterns.
Think about what "over-represents English text" actually means for topics better documented in other languages. What does the AI's pattern data on those topics look like?
8. Why is AI "confidence" not a reliable signal of accuracy?
Right. Confidence in AI output is a stylistic feature of the text, not a signal about accuracy. The model generates authoritative-sounding text because authoritative text is a prominent pattern in its training data — not because it has verified the content.
Think about calibration. The model generates what sounds plausible — and authoritative, confident writing is a very common pattern. That doesn't mean the content is accurate.
9. You use role-play prompting and ask an AI to "play a financial advisor." It gives you specific investment advice in that persona. What should you keep in mind?
Correct. Role-play prompting is stylistic — the AI generates text that sounds like that persona. It doesn't actually have the expertise, credentials, or accountability of a real financial advisor. The stakes for acting on such advice are real; the AI's simulated authority is not.
Remember: role-play changes how the AI sounds, not what it actually knows. What does "pattern-matching to a financial advisor persona" give you versus actual financial expertise?
10. The EU AI Act takes a "risk-based" approach to regulating AI. What does this mean in practice?
Right. Risk-based regulation means the rules scale with the potential for harm — AI in a video game and AI making decisions about people's employment face fundamentally different requirements under the Act.
A risk-based approach means different rules for different risk levels. Think about how AI in hiring versus AI in a game would be treated differently under such a framework.
11. An AI gives you a first response that's too long and includes irrelevant information. What's the best next step?
Exactly. Specific iterative feedback — "cut X, keep Y, aim for 100 words" — uses the context window productively. The AI can refine toward your actual needs when you tell it precisely what went wrong.
Iteration with specific feedback is the skill here. What precise instructions would help the AI understand what "too long and irrelevant" means and how to fix it?
12. A student argues that AI will make teachers unnecessary because it can explain anything. What's the strongest challenge to this argument, based on what you've learned?
Correct. The skills most durable against AI automation are precisely those involving judgment in novel situations, accountability, trust, and embodied responsiveness. A teacher does all of these; an AI generates pattern-matched explanations.
Think about the framework from Lesson 4: which human skills remain most durable? What does a teacher do that pattern-matched text can't replicate?
13. You read a headline: "AI beats radiologists at detecting cancer in scans." What question should you ask first, based on what you know?
Exactly. "Beats radiologists at detecting cancer" almost certainly refers to a narrow, well-defined pattern-recognition task on standardized data — which is exactly what AI is good at. That's meaningfully different from the full scope of a radiologist's clinical role. Headlines elide that distinction; you now know to ask for it.
Think about what AI actually excels at. Pattern recognition on labeled data is a strength — but what does that capture versus what a radiologist does in full clinical practice?
14. Why did the WGA ultimately win AI protections in their 2023 contract, while most workers in other industries didn't?
Right. Collective bargaining power is what made the WGA outcome possible. Most workers displaced or threatened by AI don't have unions with the leverage to negotiate protections — which is exactly why AI governance policy at the government level matters so much for everyone else.
Think about what was different about the WGA's situation versus, say, freelance writers or data entry workers facing AI displacement. What did the WGA have that others don't?
15. After completing this module, a classmate who hasn't taken the course asks "Is AI smart?" What's the most accurate and honest answer you could give them?
That's the honest, grounded answer. It avoids both the hype ("it's basically thinking!") and the dismissal ("it's just autocomplete") — and it points to the actual mechanism and its real implications. That nuance is exactly what separates someone who's thought carefully about this from someone who hasn't.
Think about the full picture from this module. What does AI actually do? What can it do well? Where does it fall short? The honest answer holds all of that at once.