SuperIntelligence
Home
Notes
Archive
About

SuperIntelligence

How Millions of AI Agents Work as One

Why a problem solved once rarely has to be solved again.
READ THE LATEST

Recent posts

View all
How AI Learns Values the Way Children Do
Geoffrey Hinton was right. AI is a child, and humans are its parents.
 • Dr. Craig A. Kaplan
Zuckerberg Brings the World Closer to Democratic SuperIntelligence
Mark Zuckerberg has moved remarkably close to a democratic model of SuperIntelligence.
 • Dr. Craig A. Kaplan
When AGI Becomes SuperIntelligence, Testing Is Not Enough
A Personalized SuperIntelligence can become more intelligent than the humans who designed its safeguards. Testing alone cannot guarantee its safety.
 • Dr. Craig A. Kaplan
The Four Phases of Safe AGI
A step-by-step design for training SuperIntelligence on the values of billions of people
 • Dr. Craig A. Kaplan
The Spinning Knowledge Wheel
AI must update fast-changing knowledge at the rim while human values hold still at the center.
 • Dr. Craig A. Kaplan
How AI Can Catch Danger in Real Time
When AI is unsure, it stops and asks.
 • Dr. Craig A. Kaplan
Why Training AI Is Like Testing the Brakes
AI trained by millions of people learns the situations that matter most.
 • Dr. Craig A. Kaplan
AI and the Statistics of Right and Wrong
Diversified human values protect AI safety the way a diversified portfolio protects an investor.
 • Dr. Craig A. Kaplan
How AI Should Decide When There Is No Right Answer
The Trolley Problem reveals the human values that no law can write down, and AI can learn them.
 • Dr. Craig A. Kaplan
Who Should Have the Most Influence on AI?
Two methods can combine the values of many, but the harder question is how much each voice should count.
 • Dr. Craig A. Kaplan
Who Should Teach AI Right from Wrong?
AI learns better values when more people help teach it.
 • Dr. Craig A. Kaplan
Why AI Safety Training Doesn’t Scale
The industry’s two leading safety methods fail for opposite reasons.
 • Dr. Craig A. Kaplan
SuperIntelligence
SuperIntelligence
The fastest path to superintelligence is the safest path.
Recommendations
View all 19
One Useful Thing
One Useful Thing
Ethan Mollick
Eurykosmotron
Eurykosmotron
Ben Goertzel
Artificial Intelligence Made Simple
Artificial Intelligence Made Simple
Devansh
Theory of the Game
Theory of the Game
Reid Hoffman
User's avatar
Don't Worry About the Vase
Zvi Mowshowitz
Links
iQ Company
SuperIntelligence.com
The fastest path to superintelligence is the safest path.
Craig A. Kaplan LinkedIn
iQ Company LinkedIn
iQ Studios: YouTube
X: Aligned SuperIntelligence
X: iQ Company

SuperIntelligence

The fastest path to superintelligence is the safest path.
AboutArchiveRecommendationsSitemap

SuperIntelligence

© 2026 Dr. Craig A. Kaplan · Privacy ∙ Terms ∙ Collection notice
Start your SubstackGet the app
Substack is the home for great culture