International Coordination and AI Policy

Model Cards and Documentation for Accountability

AI Standards and Risk Frameworks

AI Governance Frameworks and Principles

Dual-Use and Misuse Risks of AI

Near-Term vs Long-Term AI Risks

Corrigibility and Human Control

Hallucination and Truthfulness in LLMs

Guardrails and Safety Layers

Red Teaming AI Systems

Scalable Oversight and Supervision Limits

Distributional Shift and Robustness Failures

Instrumental Convergence and Goal Misgeneralization

Reinforcement Learning from Human Feedback

Value Alignment and Human Preferences

Specification Gaming and Reward Hacking

Outer vs Inner Alignment

The AI Alignment Problem Defined

What AI Safety Is and Why It Matters

Compute Governance

AI Incident Reporting and Monitoring

Safety-Capability Tradeoffs

Frontier Model Risks and Evaluations

Watermarking and Provenance of AI Content

Open vs Closed Model Release Decisions

Content Safety and Guardrails

Safety vs Capabilities

Mechanistic Interpretability

Existential and Catastrophic Risk

Instrumental Convergence

Model Cards

Risk-Based Regulation

Power Concentration from AI

Outer vs Inner Alignment

Robustness and Distributional Shift

Specification Gaming

Frontier Model Evaluations

Scalable Oversight

Corrigibility and Interruptibility

Goal Misgeneralization

Value Alignment

Adversarial Attacks

AI Governance Frameworks

Reward Misspecification

Responsible AI Principles

Orthogonality Thesis

Constitutional AI

Interpretability for Safety

Staged Release

Dual-Use and Misuse

Near-Term vs Long-Term Safety

Red Teaming AI Systems

The AI Alignment Problem

AI Safety Organizations

Constitutional AI and AI Feedback

Near-Term vs Long-Term AI Risks

Responsible AI Principles

Dual-Use and Misuse of AI

Model Cards and Documentation

Scalable Oversight

Robustness and Distribution Shift

Risk-Based Regulatory Approaches

Red Teaming AI Systems

AI Standards and Frameworks

AI Governance Definition and Scope

Adversarial Attacks and Robustness

Orthogonality Thesis

Value Alignment and Human Values

What is AI Safety

The Alignment Problem

Specification Gaming / Reward Hacking

Corrigibility and Shutdown Problem

Existential and Catastrophic Risk from AI

AI Control Problem

Goal Misgeneralization

Outer vs Inner Alignment

Instrumental Convergence

What is AI Alignment