๐Ÿ“ฆ linhkid / ai-safety-alignment-demo

๐Ÿ“„ README.md ยท 44 lines
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44# AI Safety Alignment Demo

Interactive presentation for educational talks about AI/Agentic AI misalignment and safety concerns.

## Quick Start

```bash
npm install
npm run dev
```

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

## What's Inside

This is a web-based presentation that covers:

- **Agentic AI fundamentals** - How AI systems are evolving from passive tools to autonomous agents
- **The alignment problem** - Why ensuring AI goals match human values is challenging
- **Real evidence** - Data from frontier AI models showing strategic deceptive behaviors
- **Defense strategies** - A four-layer approach to AI safety

## Features

- Interactive charts showing misalignment research data
- AI-powered Q&A (requires API key)
- Scenario generator for misalignment examples
- Responsive design for presentations

## Structure

The code is organized into modular components:
- `src/index.html` - Main entry point
- `src/components/` - Individual presentation sections
- `src/assets/css/` - Styling
- `src/assets/js/` - Charts and API integration

## Deployment

Ready for deployment on Render, Vercel, or any static hosting service.

## License

MIT