AI and the Law for High School Students: What a Full Day on AI Law Covers

Future Lawyers Team·Published November 18, 2026

Abstract courthouse column made of dots fading upward

AI and the law asks who is legally responsible when an AI system causes harm or produces false information. For high school students, it means learning four legal ideas (product liability, negligence, defamation and copyright), reading real cases, and practicing arguing both sides.

Legal developments in this area change quickly. This guide reflects what was publicly reported as of November 2026.

The short version

  • AI law asks who is responsible when an AI system causes harm or gets facts wrong.
  • Most AI cases lean on four older legal ideas: product liability, negligence, defamation and copyright.
  • Real cases show that lawyers stay responsible for what they file, even when AI wrote it.
  • The rules are still being written, in the EU and across the US.
  • You don't need to code to study it. You need to read carefully and argue both sides.

Why AI Law Is Worth Studying Now

Courts and lawmakers are still working out who answers for what AI tools say and do. Nobody has the final word yet. That's unusual in law, where most big questions were settled long ago, and it means students can look at a genuinely open question and form their own view.

One note before we start: this guide is education, not legal advice. It explains ideas and cases so you can think about them clearly.

Example:

Imagine a student, call her Priya, asks a chatbot about a medication dose for her younger brother. It answers confidently, and the answer is wrong. Priya wonders: if someone had acted on that, who would be responsible? That's exactly the kind of question AI law tries to answer.

The Four Legal Ideas Behind Most AI Cases

AI law isn't built from scratch. Most cases borrow ideas courts have used for a long time. These are the four the Future Lawyers Fellowship teaches.

Product liability

Product liability is the idea that a company can be responsible when a product it sells is defective and hurts someone. The open question is whether an AI tool counts as a "product" in the same way a toaster does.

Example:

A made-up health app gives a user bad medical advice, and the user is harmed after following it. Is the app a defective product, like a faulty car part? Or is it more like a book of information? Students argue both sides.

Negligence

Negligence means failing to take the care a reasonable person would take, and causing harm as a result. With AI, you can ask whether the company that built it, or the person who relied on it, failed to be careful.

Example:

A fictional delivery company uses an AI tool to plan routes and never checks its suggestions. The tool sends a driver down a closed road and there's an accident. Did the company act carelessly by trusting it blindly?

Defamation

Defamation is a false statement about a real person that damages their reputation. AI tools sometimes produce false claims about real people, which raises the question of who, if anyone, is responsible.

Example:

A fictional local baker searches her own name in a chatbot, and it wrongly says she was fined for health code violations. Customers see it. Who made the false statement: the company, the tool, or nobody?

Copyright

Copyright is the legal right creators have over their original work, like photos, music or writing. AI tools are trained on huge amounts of existing work, and creators have asked whether that use needed their permission.

Example:

A fictional photographer finds that an AI image tool can produce pictures in a style very close to hers. She never agreed to her photos being used for training. Students weigh her rights against how the tool was built.

AI Hallucinations in Court

A hallucination is when an AI tool states something false in a confident way, such as inventing a court case citation (the reference to where a decision can be found). The fake case can look completely real.

That's what happened in Mata v. Avianca, a case in a New York federal court in 2023. Lawyers filed a court document citing cases that an AI chatbot had invented. The court sanctioned the lawyers, meaning it penalized them. The lesson is simple: check every citation in a primary source, the actual decision itself.

Example:

A student uses an AI tool to help brainstorm a research essay on free speech. Before submitting, he looks up every source it suggested. Two don't exist, so he replaces them with real ones he found himself.

Real Cases to Know

CaseWhat it was aboutWhy it matters
Mata v. AviancaLawyers in a New York federal court filed a document citing cases an AI chatbot had invented (2023).The court sanctioned the lawyers. Using AI doesn't move responsibility away from the person who files.
Walters v. OpenAIA radio host sued OpenAI in Georgia state court for defamation after ChatGPT falsely said he had been accused of embezzlement.The court granted summary judgment for OpenAI in 2025, so the claim did not succeed. It shows how hard defamation can be to prove against an AI company.
Getty Images v. Stability AIGetty Images sued Stability AI in 2023 over the use of its images to train an AI image generator.It raises the copyright question of training data. The cases are in different countries and have developed since, so check current news.

Legal outcomes in this area change quickly. Check current reporting for the latest.

Example:

A student picks Mata v. Avianca and writes a one-page argument: should lawyers have to tell the court when they've used AI? She uses the case as her starting fact, then argues one side and answers the other.

What the Rules Look Like Today

The European Union has a broad AI law, the EU AI Act, which applies in stages. It's one of the first attempts to set rules for AI across many countries at once.

In the US, there is no single federal AI law. Instead there's a changing mix of federal guidance and state laws. On both sides of the Atlantic, the rules keep changing, which is part of what makes this area worth following.

Five Rules for Using AI Safely

  1. Check every case yourself in a primary source, meaning the actual court decision, not a summary.
  2. AI predicts what legal writing sounds like. It doesn't actually know the law.
  3. Lawyers are still responsible for their work, even when they use AI to help.
  4. Advice from an AI tool is not legal advice.
  5. Courts are starting to ask people to disclose when they've used AI, so learn the rules wherever you practice.

What a Full AI Law Day Looks Like in the Fellowship

In the Future Lawyers Fellowship, Weekend 3 Saturday is given to AI law. It starts with an introduction to who is responsible for what an AI generates, then moves to how to sue an AI company and to AI hallucinations. Students take part in an AI liability simulation, then finish with a workshop to polish their final project and prepare for moot court. You can see the outline in the AI and the law section of our homepage.

Example:

Here's a classroom exercise: an AI gives bad medical advice and someone is harmed. Who do you sue? Half the room argues for the company, half for the user, and each side has to name which legal idea they're relying on.

How to Start Exploring AI Law on Your Own

For the longer road, read our guide on how to become a lawyer.

FAQ

What is AI law?

AI law is the set of legal questions about who is responsible when an AI system causes harm or gets something wrong. It draws on older ideas like negligence, defamation and copyright, plus new rules written specifically for AI.

Can you sue an AI company?

People have tried, using several legal ideas such as defamation, product liability and copyright. Outcomes have varied, and the law is still unsettled. This is general education, not legal advice.

What are AI hallucinations?

A hallucination is when an AI tool states something false in a confident way, like inventing a court case that doesn't exist. The answer can look polished and still be wrong, so every fact and citation needs checking.

What happened in Mata v. Avianca?

In 2023, lawyers in a New York federal court filed a document citing cases that an AI chatbot had invented. The court sanctioned the lawyers, which means it penalized them for what they filed.

Do I need to know coding to study AI and the law?

Not for the basics. The questions are about responsibility and evidence, not programming. Curiosity and careful reading matter more than code.

Curious how an AI law session feels?

Join a free live masterclass first, or apply to the Future Lawyers Fellowship with Harvard student mentors.

Next Fellowship cohort: 6 March 2027 · Early Bird $499 USD · Regular $699 USD per student