Human Rating - AI Companion Rating Platform

A collaborative, wiki-style platform for evaluating and rating AI systems and companions with a focus on human-centered metrics and transparent scoring.

Overview

Human Rating is a platform designed to provide comprehensive, human-centered evaluations of AI systems and companions. It integrates a Stack Overflow-inspired reputation system, community-driven content, and a structured, multi-dimensional scoring system.

Features

• Humane Tech Scorecard: Evaluate AI systems based on how they make users feel:

  • Cared for: Emotional support and empathy
  • Present: Attentiveness and engagement
  • Fulfilled: Accomplishment and satisfaction
  • Connected: Sense of belonging and connection

• Wiki-Style Collaboration: All AI entries and reviews are collaboratively editable, with version control and transparent edit histories.

• Reputation System: Users earn reputation through upvotes, accepted contributions, and approved edits, unlocking more privileges as their reputation increases.

• Structured Impact Assessment: Comprehensive evaluation framework covering intended use, technical capabilities, limitations, risks, and team information.

• Comparative Testing: Side-by-side comparisons of AI responses to understand relative strengths and weaknesses.

• Red Team Findings: Community members can probe AIs for harmful or biased outputs, reporting findings for transparency and improvement.

Technology Stack

• Frontend: Next.js, React, TypeScript, Tailwind CSS
• Backend: Next.js API routes
• Database: Prisma ORM with SQLite (can be configured for other databases)
• Authentication: (Placeholder for future implementation)

Getting Started

Prerequisites

• Node.js 18 or higher
• npm or yarn

Installation

1. Clone the repository

git clone https://github.com/yourusername/human-rating.git
cd human-rating

2. Install dependencies

npm install
# or
yarn install

3. Set up the database

npx prisma migrate dev --name init

4. Start the development server

npm run dev
# or
yarn dev

5. Open http://localhost:3000 in your browser

Project Structure

• /src/app - Next.js app router pages and layouts
• /src/components - Reusable React components
• /src/lib - Utilities and library code
• /src/models - Data models and types
• /prisma - Database schema and migrations

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

This project was inspired by the need for more human-centered evaluation frameworks for AI systems, particularly as AI companions become more integrated into our daily lives.

Deployment

Deploying to Vercel

1. Push your code to a Git repository (GitHub, GitLab, or Bitbucket)

2. Sign up or log in to Vercel

3. Click "Add New" > "Project"

4. Select your Git repository

5. Configure the project:

   • Framework Preset: Next.js
   • Build Command: Leave as default (uses the vercel.json configuration)
   • Output Directory: Leave as default (.next)

6. Add the following environment variables:

   • DATABASE_URL: Your PostgreSQL connection string
   • NEXTAUTH_URL: The URL of your deployed site
   • NEXTAUTH_SECRET: A random string used for encryption

7. Click "Deploy"

Database Setup for Production

For production, the application uses PostgreSQL instead of SQLite. You can use Vercel Postgres or any other PostgreSQL provider.

1. Set up a PostgreSQL database (Vercel Postgres, Supabase, Railway, etc.)

2. Get your PostgreSQL connection string

3. Set it as the DATABASE_URL environment variable in your Vercel project settings

4. The database migration will run automatically during deployment