Ethical Decision Explorer

An interactive web application that helps explore ethical questions through decision trees, powered by LLMs.

Features

• Interactive decision tree visualization
• Multiple ethical theories/frameworks support:
  • Utilitarianism
  • Deontological Ethics
  • Virtue Ethics
  • Contractualism
  • Care Ethics
• Real-time AI-powered decision tree generation
• Responsive and modern UI with Nuxt UI
• Beautiful animated tree layouts with Vue Flow

Analysis Methods

Maximize Expected Choice-Worthiness (MEC)

When faced with an ethical decision, the app helps you weigh different moral perspectives by calculating Expected Choice-Worthiness:

• Ethical Theory credence: Assigns probability weights (0-100%) to different ethical theories based on how likely they are to be true
• Choice-worthiness: Evaluates the moral value of each choice (-1000 to 1000) based on how well it aligns with different theories
• Expected value calculation: Multiplies credence by choice-worthiness to produce an overall ethical score
• Multi-theory analysis: Incorporates perspectives from:
  • Utilitarianism (maximizing overall good)
  • Deontology (moral rules and duties)
  • Virtue Ethics (character development)
  • Contractualism (justice and mutual agreement)
  • Care Ethics (relationships and care)

Moral Parliament Deliberation

Complements the quantitative analysis with a deliberative approach that simulates stakeholder perspectives:

• Stakeholder representation: For each stakeholder:

  • Identity and role in the decision
  • Ethical perspective and viewpoint
  • Support score (0-10) indicating position strength
  • Detailed reasoning for their stance

• Deliberation outcome:

  • Consensus score (0-10) measuring stakeholder agreement
  • Analysis of competing viewpoints
  • Potential paths to resolution
  • Final synthesis with confidence level

By combining MEC with moral parliament deliberation, you can explore how different ethical theories compete and overlap, hear from various stakeholders, and arrive at well-reasoned decisions with confidence in your ethical reasoning.

Technologies

• Nuxt.js 3
• Vue 3
• TypeScript
• Vue Flow
• Nuxt UI
• Tailwind CSS
• Google Gemini API

Prerequisites

• Node.js 20 or later
• A Google Gemini API key

Setup

1. Clone the repository:

git clone https://github.com/Katsari/ethical-decision-explorer-ai.git
cd ethical-decision-explorer-ai

2. Install dependencies:

npm install

3. Start the development server:

npm run dev

4. Open your browser and navigate to http://localhost:3000

License

MIT