Trustworthy AI — Streamlit Overview Demo (for politicians & management)

This Streamlit app is an overview demo that explains:

• What Trustworthy AI is
• Why it matters (public safety, economic, trust/legitimacy)
• A simple interactive mini-demo showing how safeguards change risk
• Short failure stories + a practical roadmap

Quick start

1) Create a virtualenv (recommended)

python -m venv .venv
# macOS/Linux
source .venv/bin/activate
# Windows
# .venv\Scripts\activate

2) Install dependencies

pip install -r requirements.txt

3) Run the app

streamlit run app.py

Notes

• The interactive demo uses a small example dataset and a lightweight simulated scoring model (no heavy ML dependencies) to keep the demo easy to run and easy to understand.
• You can replace data/sample_cases.csv with your own domain examples later.

Structure

• app.py — Home / navigation
• pages/ — Streamlit multipage content
• trust_utils.py — shared scoring + helper functions
• data/sample_cases.csv — small example dataset

License

MIT

Story mode

Open Interactive mini-demo (story mode) and click through: Reliable → Safe → Fair → Transparent → Accountable.