Research organization working on AI safety and alignment research.
Related Wiki Pages
Top Related Pages
AI Alignment
Technical approaches to ensuring AI systems pursue intended goals and remain aligned with human values throughout training and deployment. Current ...
Why Alignment Might Be Hard
AI alignment faces fundamental challenges: specification problems (value complexity, Goodhart's Law), inner alignment failures (mesa-optimization, ...
AI Alignment Research Agendas
Analysis of major AI safety research agendas comparing approaches from Anthropic (\$100M+ annual safety budget, 37-39% team growth), DeepMind (30-5...
Alignment Robustness Trajectory Model
This model analyzes how alignment robustness changes with capability scaling. It estimates current techniques maintain 50-65% robustness at GPT-4 l...
Value Learning
Training AI systems to infer and adopt human values from observation and interaction