Skip to content

Ethical Dimensions of AI

Run the Ethical Dimensions of AI MicroSim Fullscreen
Edit in the p5.js Editor

About This MicroSim

A radar chart or multi-axis diagram allows students to see that AI ethics is not a single issue but a multi-dimensional space of competing values, where improving one dimension may create tension with another. Students can use sliders to adjust ethical constraints within transparency, fairness, accountability, privacy, and autonomy.

Lesson Plan

Grade Level

9-12 (High School / IB TOK)

Duration

15-20 minutes

Prerequisites

  • Basic understanding of how machine learning systems use data.
  • Foundational awareness of why ethical guidelines matter for technology.

Learning Objectives

  • Appraise the ethical trade-offs involved in AI deployment by weighing competing values such as efficiency, transparency, fairness, and autonomy.

Activities

  1. Exploration (5 min): Let students manipulate the sliders across the 5 axes (Transparency, Fairness, Accountability, Privacy, and Autonomy) for the default "Social Media Algorithm" scenario. Ask them to observe what happens to the resulting polygon shape.
  2. Guided Practice (10 min): Use the dropdown to select the "Medical Diagnosis AI" and the "Criminal Sentencing AI" presets. Have students discuss in pairs why certain values (like Accountability or Privacy) drastically decrease or increase depending on the scenario context.
  3. Assessment (5 min): Instruct students to configure the axes to create what they believe is the "Ideal Ethical Profile." Have them explain via a short journal entry which value they heavily prioritized and what value they sacrificed to achieve that.

Assessment

  • Participation in guided practice scenarios.
  • Written or verbal reflection analyzing the trade-offs they discovered in their ideal AI ethical profile.

Quiz

Test your understanding of the ethical dimensions of AI with this review question.

1. When configuring an AI system for medical diagnosis, maximizing patient "Privacy" typically results in a direct trade-off with which other dimension?

  1. Hardware efficiency
  2. Transparency of the system's learning algorithm
  3. Accountability
  4. Artificial autonomy
Show Answer

The correct answer is B. Increasing privacy typically limits the amount of raw, open data that can be analyzed publicly, which directly forces the creators to reduce transparency in how the AI model parses specific individual traits to make decisions.

Concept Tested: Ethical Trade-Offs in AI