This simple app, created on a shoestring (fixed costs of <$10/month) with Django on PythonAnywhere, Supabase PostgreSQL with Pgvector extension (as the vector database), and Huggingface Pro inference APIs (for LLMs), is designed to help one learn about AI Governance, Regulations, and Risks via question answering and flashcards. It extracts information from the list of resources at AI Governance, Regulations and Risks Database and uses the information as a knowledge base with a Mixtral-8x7B-Instruct-v0.1 to generate answers and flashcards for questions. Embeddings are generated using bge-small-en-v1.5.

Features include:

• Question answering
• Flashcards for related questions and answers
• User feedback system for continuous improvement of knowledge base

The answers should get better over time, as the questions and answers that are rated as good are added to the flashcard QA database and used as context in subsequent queries. The usual disclaimer relating to LLMs applies. Do NOT trust the outputs and always verify.