Repositorio de estudio. Study repository.
50 recursos. Busca por nombre, o filtra por área, nivel y tipo. Seguimos sumando recursos: si conoces alguno que debería estar aquí, escríbenos a hola@chais.cl. 50 entries. Search by name, or filter by area, level and kind. We keep adding resources: if you know one that should be here, email us at hola@chais.cl.
EntradaGetting started 18
Por dónde empezar si no has leído nada del tema.Where to start if you have not read anything on the subject.
FundacionalFoundational 8
- BlueDot Impact — cursos actuales
- Robert Miles AI Safety (canal de YouTube)
- Preventing an AI-related catastrophe (80,000 Hours)
- AI Safety Info / Stampy
- AI Safety for Fleshy Humans (Nicky Case)
- AI Safety, Ethics and Society (Hendrycks) — libro de texto
- AI Safety Atlas
- Intro to ML Safety (CAIS — curso completo)
PrácticoPractical 2
Joya OcultaHidden gem 8
TécnicoTechnical 11
Alineamiento, interpretabilidad y evaluación de modelos.Alignment, interpretability and model evaluation.
FundacionalFoundational 8
- Concrete Problems in AI Safety (Amodei et al., 2016)
- Zoom In: An Introduction to Circuits (Distill)
- The Alignment Problem from a Deep Learning Perspective (Ngo, Chan, Mindermann)
- Constitutional AI: Harmlessness from AI Feedback
- Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
- Agentic Misalignment (Anthropic, 2025)
- The case for ensuring that powerful AIs are controlled (Redwood)
- Alignment Faking in Large Language Models
PrácticoPractical 2
Joya OcultaHidden gem 1
GobernanzaGovernance 10
Regulación, política pública y estándares.Regulation, public policy and standards.
FundacionalFoundational 5
PrácticoPractical 3
Joya OcultaHidden gem 2
Bio-riesgoBiorisk 1
El cruce entre inteligencia artificial y riesgo biológico.Where artificial intelligence meets biological risk.
FundacionalFoundational 1
EnsayosEssays 10
Los textos que le dieron forma a la discusión.The texts that shaped the discussion.