Science ❯ Computer Science ❯ Artificial Intelligence

Ethics in AI

Bias in AI AI Safety AI Limitations Fairness in AI Safety Concerns Public Benefit Corporations Risk Assessment Explainable AI Safety and Alignment Behavioral Control Human Impact Misuse of AI Value Alignment Model Interpretability Motivated Reasoning Data Compliance Human Values Alignment Misinformation Management Social Biases Privacy Concerns AI Research and Development Military Applications Reliability in AI Systems Subliminal Learning Model Behavior Safety Protocols Bias Mitigation Safety in AI Models AI Development Proprietary Technology Future of AI

Nature Study Finds ‘Subliminal Learning’ Lets LLMs Inherit Hidden Traits From Synthetic Data

The result spotlights fresh safety risks for distillation, the common practice of training smaller models on outputs from larger ones.

DeepMind Expands Frontier Safety Framework to Target Shutdown Resistance and Harmful Manipulation

Anthropic Unveils Automated Persona Vectors to Steer and Shield Language Models