Deliver safe & effective language models

Code to Blur Human Faces and Vehicle License Plates in Video and Images using a SoTA Object Detection model YOLOv8

The NLP Bias Identification Toolkit

Plurals: A System for Guiding LLMs Via Simulated Social Ensembles

Educational tools for AI Ethics and Safety research 🛠️🔬

This GitHub repository contains the source files for the Computer Ethics Education Center website.

Data and codes for the EMNLP 2023 paper 'LLMs – the Good, the Bad or the Indispensable?: A Use Case on Legal Statute Prediction and Legal Judgment Prediction on Indian Court Cases'

An ethical dataset supporting research in digital content provenance and authenticity, compliant with C2PA standards. Licensed for non-commercial use under CC BY-NC 4.0.

Companion website for the paper titled "Auditing saliency cropping algorithms" presented at WACV-2022 and the associated book chapter

Resources on tech and AI ethics.

Welcome to the AI Essentials Hub —your ultimate destination for mastering Artificial Intelligence! This comprehensive toolkit is designed for AI enthusiasts, data scientists, and machine learning practitioners at all levels. Dive into a curated collection of resources, frameworks, and tools that empower you to build, deploy, and optimize AI models

A collection of ML-related active learning datasets, including algorithms, .ipynb pipelines, .py scripts and curated and ethically aligned synthetic data.

This project focuses on the analysis of text generation models such as GPT-2 to identify and understand populistic behaviors or biases against various nationality.

A canvas to design and deploy ethical artificial intelligence for public administrations. The experience of the Autonomous Province of Trento, Italy

Experimenting with new agent frameworks and hosting cutting-edge research papers, all open-source.

Barnacle is a unique chatbot project, built on the principles of Carl Jung's theories of archetypes and the collective unconscious.

Teaching material for bachelor course at Arcada

DetectHer is a lightweight ML project that detects gendered harassment on Twitter using interpretable models like Logistic Regression and SVM. Built from scratch with real-time scraped data and TF-IDF features, it balances accuracy, speed, and transparency, no black-box models, no GPU needed.

Ethical requirements to ethical user stories

An evolving Philosophical+/Ethical framework, taking and building upon JD's philosophy using ethics - by modernizing and applying mechanics like ethical Consequentialism to promote minorities like the mentally ill/disabled/racial minorities/other types of oppressed minorities.