AI Safety Compendium

Home

❯

concepts

Folder: concepts

101 items under this folder.

05 May 2026
AI Alignment
- alignment-techniques
05 May 2026
AI Control
- alignment-techniques
05 May 2026
AI Governance
- governance
05 May 2026
AI Risk Arguments
- risk-models
05 May 2026
AI Safety
- field-infrastructure
05 May 2026
AI Takeover Scenarios
- risk-models
05 May 2026
Capability Evaluations
- evaluations
05 May 2026
Constitutional AI (RLAIF)
- alignment-techniques
05 May 2026
Dangerous Capabilities
- evaluations
- risk-models
05 May 2026
Existential Risk
- risk-models
05 May 2026
Goal Misgeneralization
- risk-models
05 May 2026
Goodhart's Law
- risk-models
05 May 2026
Instrumental Convergence
- risk-models
05 May 2026
Interpretability
- interpretability
05 May 2026
Iterative Amplification
- alignment-techniques
05 May 2026
Mesa-Optimization
- risk-models
05 May 2026
Model Organisms of Misalignment
- evaluations
05 May 2026
Outer vs. Inner Alignment
- risk-models
- alignment-techniques
05 May 2026
Power Seeking
- risk-models
05 May 2026
Responsible Scaling Policy
- governance
05 May 2026
Reward Hacking
- risk-models
05 May 2026
Reinforcement Learning from Human Feedback (RLHF)
- alignment-techniques
05 May 2026
Scheming
- risk-models
05 May 2026
Situational Awareness
- risk-models
- evaluations
05 May 2026
Specification Gaming
- risk-models
05 May 2026
Superalignment
- alignment-techniques
05 May 2026
Takeoff Dynamics
- forecasting
29 Apr 2026
AI Red Lines
29 Apr 2026
AI Safety Levels (ASL)
29 Apr 2026
AI Safety via Debate
29 Apr 2026
Brussels Effect
29 Apr 2026
Compute Governance
29 Apr 2026
Control Evaluations
29 Apr 2026
Data Governance
29 Apr 2026
Evaluated Properties
29 Apr 2026
Evaluation Design
29 Apr 2026
Evaluation Frameworks
29 Apr 2026
Evaluation Limitations
29 Apr 2026
Evaluation Techniques
29 Apr 2026
Frontier Safety Frameworks
29 Apr 2026
Goal-Directedness
29 Apr 2026
Governance Architectures
29 Apr 2026
Governance Problems
29 Apr 2026
If-Then Commitments
29 Apr 2026
Inverse Reinforcement Learning (and CIRL)
29 Apr 2026
Learning Dynamics
29 Apr 2026
Multi-Objective Generalization
29 Apr 2026
National AI Governance
29 Apr 2026
Process Oversight
29 Apr 2026
Propensity Evaluations
29 Apr 2026
Reward Tampering (and Wireheading)
29 Apr 2026
Task Decomposition
29 Apr 2026
Verification vs. Generation
29 Apr 2026
The Waluigi Effect
28 Apr 2026
AGI Definitions and Thresholds
28 Apr 2026
AI Autonomy Levels
28 Apr 2026
AI Risk Management
28 Apr 2026
AI Safety Culture
28 Apr 2026
Alignment to Whom
28 Apr 2026
Alternative Risk Categories
28 Apr 2026
ASI Safety Strategies
28 Apr 2026
Autonomous Replication
28 Apr 2026
Circuit Breakers (AI Safety)
28 Apr 2026
Coherent Extrapolated Volition (and CAV / CBV)
28 Apr 2026
CoT Monitoring (Technique)
28 Apr 2026
Defense in Depth
28 Apr 2026
Defensive Acceleration (d/acc)
28 Apr 2026
Effective Compute
28 Apr 2026
Enfeeblement
28 Apr 2026
Epistemic Erosion
28 Apr 2026
Foundation Models
28 Apr 2026
Global AI Moratorium
28 Apr 2026
Machine Unlearning
28 Apr 2026
Mass Unemployment
28 Apr 2026
Misuse Prevention Strategies
28 Apr 2026
Mutual Assured AI Malfunction (MAIM)
28 Apr 2026
P(doom)
28 Apr 2026
Pivotal Act
28 Apr 2026
Risk Amplifiers
28 Apr 2026
Risk Decomposition
28 Apr 2026
Scaling Laws
28 Apr 2026
Socio-Technical Strategies
28 Apr 2026
Systemic Risks
28 Apr 2026
Transformative AI
27 Apr 2026
Deceptive Alignment
- risk-models
27 Apr 2026
Mechanistic Interpretability
- interpretability
27 Apr 2026
Robustness
27 Apr 2026
Scalable Oversight
- alignment-techniques
23 Apr 2026
Biosecurity
23 Apr 2026
Near-Term Harms vs. Long-Term X-Risk
21 Apr 2026
AI Agents
21 Apr 2026
AI Military Applications
21 Apr 2026
AI Population Explosion
21 Apr 2026
Autonomous Weapons
21 Apr 2026
Distribution Shift
21 Apr 2026
Reward Learning
16 Apr 2026
Intelligence Explosion
08 Apr 2026
Differential Development
08 Apr 2026
Information Security
08 Apr 2026
Stable Totalitarianism
08 Apr 2026
Value Lock-In

Suggest a source
Connect
Overview
About (proof of concept)
Email feedback
Made by IT for Humanity

Explorer

Folder: concepts