Legal and Ethics in AI

As AI becomes integrated into our lives, its ethical and legal implications become paramount. This field focuses on ensuring intelligent systems are developed and deployed responsibly, addressing critical issues like fairness, bias, transparency, and accountability. My approach is to build AI that is not only powerful but also trustworthy and aligned with human values.

Studied Legislation:

nLPD: The new Swiss Data Protection Act, which aligns with the GDPR and sets standards for data processing and privacy in Switzerland.

GDPR: The General Data Protection Regulation, a comprehensive data protection law in the EU that sets guidelines for the collection and processing of personal information.

EU AI Act: A comprehensive regulatory framework for AI in the European Union, focusing on risk-based categorization of AI systems and ensuring compliance with ethical standards.

Key Technologies & Frameworks:

Explainable AI (XAI) (SHAP, LIME): Libraries for interpreting “black box” model predictions, making their decisions transparent and understandable to humans.

Fairness & Bias Detection (Fairlearn, AIF360): Toolkits for auditing machine learning models for biases and applying mitigation algorithms to promote equitable outcomes.

Differential Privacy (Google’s DP Library, PyDP): Frameworks for training models on sensitive data while providing mathematical guarantees of individual privacy.

Key Projects:

Scalable oversight: A presentation that explores the modern techniques to ke you AI under control.