AI UNIT 3 • FOUNDATIONS • FREE

Who Trains the AI?

An AI is not neutral. It carries the choices and blind spots of the people and data behind it. In this unit you'll become an investigator, hunting down where the machine gets Native peoples wrong, or leaves them out entirely, and tracing it back to its source.

🪶 5 Stages
🌱 Foundations · No coding
🎯 30-45 min per stage
💬 Guided AI Sandbox

The Big Idea

Unit 2 showed you that a machine copies whatever examples it's fed. This unit follows that idea somewhere important: into bias and erasure. When an AI describes Native peoples with tired stereotypes, or talks about your nation only in the past tense, or has almost nothing to say at all, that isn't the machine having an opinion. It's the machine faithfully repeating what people put into it.

There are two harms to watch for. Misrepresentation is when the machine describes you, but wrongly, flattening hundreds of distinct nations into one costume. Erasure is quieter and just as harmful: being left out, treated as gone, or never mentioned at all. Both trace back to who built the machine and whose voices they used.

This is critical media literacy, the skill of seeing the choices behind what a screen shows you. And you're bringing something to it that a lot of people can't: you know what accurate looks like, because it's your own community. That makes you exactly the right person to hold the machine accountable.

💬 How this unit works. Same guided sandbox: an Ask the AI panel and a Field Notes panel, no coding. This time you're an investigator, gathering evidence of bias and erasure and tracing where it came from.

By the end of this unit, you'll be able to say "I can..."

  • Explain what bias in an AI system means, and where it comes from
  • Find and document real examples of Native misrepresentation or erasure in AI output
  • Tell the difference between bias and a simple factual mistake
  • Explain what "erasure" is, and why being left out is its own harm
  • Argue, using CARE principles, what accountable AI would require

What You'll Make

An investigator's Field Notes: documented evidence of how an AI misrepresents and erases Native peoples, with your own examples.

You'll collect stereotypes, past-tense framing, and flat-out gaps, then trace each one back toward its source in the people and data that trained the machine. You'll finish by writing a short argument for what an accountable AI would have to do differently.

It's evidence you gathered, about your own community, in your own words. That's the foundation the data-sovereignty and ethics units are built on.

Your learning path

The 5 Stages

STAGE 1 What Is Bias in a Machine? It's inherited, not invented STAGE 2 Erasure: When You're Not There The harm of being left out STAGE 3 When You're There but Wrong Stereotypes and flattening STAGE 4 Follow It Back Trace bias to its source STAGE 5 Your Field Notes What accountable AI would require
Let's Begin →

← Back to All AI Units

Sources

The bias and media-literacy concepts follow the leading K-12 AI education frameworks, and the framing of representation and erasure follows Indigenous data governance scholarship. We prioritize the national bodies that write AI standards, Native-led organizations, and peer-reviewed research.

AI Literacy & Media Literacy

Indigenous Data, Representation & Sovereignty

Digital Citizenship

Educational Standards

This unit aligns with the national AI education framework (AI4K12), Indigenous data governance principles (CARE), and national and state computer science, media literacy, and social studies standards. Open any section to see how the unit meets it.

  • Big Idea 5, Societal Impact (Grades 5-12): AI can impact society in both positive and negative ways, including bias. (The entire unit is a hands-on investigation of a specific, real societal harm: how AI misrepresents and erases Native peoples.)
  • Big Idea 3, Learning (Grades 6-12): Computers learn from data. (Students trace bias in AI output back to biased or missing training data, connecting cause to effect.)
  • Big Idea 4, Natural Interaction (Grades 5-12): Intelligent agents interact with humans in many ways. (Students probe the AI conversationally to surface bias that only shows up across many prompts.)
  • Ethics (E): Indigenous peoples' rights and wellbeing are the primary concern across the data life cycle. (Students document harm, misrepresentation and erasure, rather than repeat it, and argue for an ethic of accurate, respectful representation.)
  • Collective Benefit (C): Data systems should benefit Indigenous peoples. (Students ask who is harmed and who benefits when an AI represents their community poorly.)
  • Authority to Control (A): Indigenous peoples have rights in how they are represented. (The unit frames accurate self-representation as a right the machine currently ignores.)
  • OSEU 2, Sovereignty: Tribal nations are sovereign and have the right to self-representation. (Erasure and misrepresentation are examined as failures to honor a nation's right to represent itself.)
  • OSEU 7, Learning & Identity: Understanding and language shape identity and worldview. (Students consider how a machine's distorted picture of their people can shape how others, and young Native people, see them.)
  • OSEU 1, Distinct Tribal Nations: Indigenous peoples are organized into many distinct nations. (Students specifically catch the AI flattening hundreds of distinct nations into a single stereotype.)
  • CSTA 2-IC-21 (Grades 6-8): Discuss issues of bias and accessibility in the design of existing technologies. (This unit is a sustained, evidence-based investigation of bias in a real AI system.)
  • CSTA 2-IC-20 (Grades 6-8): Compare tradeoffs associated with computing technologies that affect people's lives. (Students weigh the real cost to a community when an AI erases or stereotypes it.)
  • ISTE 1.2, Digital Citizen: Students act in ways that are safe, legal, and ethical. (Students practice the ethical habit of naming harm and refusing to repeat it.)
  • ISTE 1.3, Knowledge Constructor: Students critically evaluate the accuracy and perspective of information. (Students evaluate AI output for accuracy and for whose perspective it reflects.)
  • MN Media Arts 2.6.5.10.1 / 2.7.5.10.1 / 2.8.5.10.1 (Grades 6-8): Media artworks are influenced by cultural and historical contexts, including MN American Indian Tribes and communities. (Students analyze how AI-generated media misrepresents or erases MN Tribal nations.)
  • MN ELA, Analyzing Perspective & Evidence (Grades 6-8): Analyze how an author's or source's perspective shapes a text. (Students identify whose perspective an AI's answers reflect, and whose they leave out.)
  • Note for RF: verify exact MN codes against the current published standards before publishing.
  • ND CS 6.S.1 / 7.S.1 / 8.S.1 (Grades 6-8): Examine positive and negative impacts of technology, including equitable access. (Students investigate a concrete negative impact: biased and erasing AI output about Native peoples.)
  • ND Indigenous Language Standard 2.1 (All Grades): Investigate and reflect on the practices and perspectives of the cultures studied. (Students contrast the machine's flattened picture with the real distinctiveness of their nation.)
  • Note for RF: verify exact ND CS codes against the current published frameworks before publishing.
  • OSEU Standard 2 (All Grades): Understand sovereignty, including the right to self-representation. (Erasure and stereotyping are framed as failures to honor that right.)
  • OSEU Standard 1 (All Grades): Understand the distinctness of the Oceti Sakowin and other nations. (Students catch the AI collapsing distinct nations into one.)
  • SD CS 6-8.IC.01 (Grades 6-8): Compare tradeoffs of computing technologies affecting urban, rural, and reservation communities. (Students weigh the cost to Native communities of biased AI representation.)
  • Note for RF: verify exact SD CS codes and OSEU numbering against the current published frameworks before publishing.