AI Ethics & Safety

Fairness, bias, alignment, interpretability, and building responsible AI systems.

Fairness & Bias

Detecting and mitigating bias in data, models, and AI decision-making.

Interpretability & Explainability

SHAP, LIME, attention visualization, and making AI decisions transparent.

Aligning AI systems with human values, goals, and intentions.

Regulation, policy, standards, and frameworks for responsible AI deployment.

Adversarial Robustness

Adversarial attacks, defenses, certified robustness, and model security.

Probing model vulnerabilities, safety testing, and adversarial evaluation.

Social Impact of AI

Job displacement, inequality, surveillance, and societal effects of AI systems.