Skip to content
LongtermWiki

Comparison Tables

Interactive comparison tables for exploring AI safety, architectures, and risks. Click any table below to view it in detail.

View Diagrams →

Safety & Alignment

Safety Approaches
Table
Evaluate safety research on effectiveness vs capability uplift. Which techniques actually make the world safer?

12 cols × 40 rows

Safety Generalizability
Table
Which safety approaches generalize across different AI architectures and deployment patterns?

8 cols × 35 rows

Safety × Architecture Matrix
Matrix
Matrix view showing compatibility between safety approaches and architecture scenarios.

Intelligence Paradigms

Architecture Scenarios
Table
Compare deployment patterns and base architectures. Safety outlook, research tractability, key properties.

15 cols × 12 rows

Deployment Architectures
Table
Compare how AI systems are deployed: API access, on-device, edge computing, etc.

10 cols × 8 rows

Risks

Accident Risks
Table
Compare accident and misalignment risks by severity, likelihood, and detectability.

10 cols × 15 rows

Evaluations

Evaluation Types
Table
Compare different evaluation methodologies for AI systems.

8 cols × 20 rows

AI Transition Model

AI Transition Model Parameters
Table
All parameters in the AI Transition Model with ratings for changeability, uncertainty, and impact.

5 cols × 50 rows

See also: