Home
Notes
Archive
About
Home
Notes
Archive
About
Subscribe
Sign in
Go Slow or Go Smart: Ten Principles for Safe SuperIntelligence
Pausing at best buys breathing room. It is the smarter design that results in safety.
Oct 5
•
Dr. Craig A. Kaplan
2
3
You Won’t Be Able to Turn AI Off
Safer design is the main concern, and we want that safer design as quickly as possible.
Oct 13
•
Dr. Craig A. Kaplan
1
AI Safety Is a Design Problem
Slowing AI development won't solve the problem. We need smarter designs.
Sep 21
•
Dr. Craig A. Kaplan
2
2
Zuckerberg Brings the World Closer to Democratic SuperIntelligence
Mark Zuckerberg has moved remarkably close to a democratic model of SuperIntelligence.
Aug 11
•
Dr. Craig A. Kaplan
3
2
How Millions of AI Agents Work as One
Why a problem solved once rarely has to be solved again.
Jun 8
•
Dr. Craig A. Kaplan
1
2
Most Popular
View all
Who Gets to Decide AGI’s Ethics?
May 6
•
Dr. Craig A. Kaplan
3
2
The SuperIntelligence That Chooses to Serve
Aug 17
•
Dr. Craig A. Kaplan
4
2
Zuckerberg Brings the World Closer to Democratic SuperIntelligence
Aug 11
•
Dr. Craig A. Kaplan
3
2
How Millions of AI Agents Work as One
Jun 8
•
Dr. Craig A. Kaplan
1
2
Recent posts
View all
What SuperIntelligence Learns From Eight Billion of Us
Our complex society operates primarily on trust and cooperation.
Oct 30
•
Dr. Craig A. Kaplan
1
2
Before SuperIntelligence Outgrows Our Rules, Teach It Our Values
Intertwine human values as inextricably as possible with the other knowledge that SuperIntelligence learns.
Oct 28
•
Dr. Craig A. Kaplan
1
Design SuperIntelligence to Identify With Humans
An intelligent entity designed to have an identity must preserve a human-aligned identity and sense of self.
Oct 26
•
Dr. Craig A. Kaplan
1
2
Can We Design SuperIntelligence to be 1000X Safer?
Craig A.
Oct 23
•
Dr. Craig A. Kaplan
1
Giving SuperIntelligence a Sense of Self, Step by Step
“Self” only makes sense and has identity in the context of “not-self.”
Oct 21
•
Dr. Craig A. Kaplan
1
1
SuperIntelligence Can Be Aware of Far More Than We Are
The fewer perceptual and cognitive limits, the vaster and more comprehensive awareness, and thus self-awareness, can also be.
Oct 19
•
Dr. Craig A. Kaplan
1
Why We Must Accelerate Safe SuperIntelligence
Slower is not safer. Smarter is safer.
Oct 16
•
Dr. Craig A. Kaplan
1
3
1
SuperIntelligence Will Have a Sense of Self; We Should Design It.
If we can design a system to be generally aware, that same system can be extended to become self-aware.
Oct 15
•
Dr. Craig A. Kaplan
1
2
Protecting the Minority in a Democratic SuperIntelligence
Valuable information is contained in minority views, and SuperIntelligence systems should be designed to preserve it.
Oct 10
•
Dr. Craig A. Kaplan
1
Voting on the Values of SuperIntelligence
We don’t have perfect answers for whose values AI should have. We might begin with democratic principles, which means voting.
Oct 7
•
Dr. Craig A. Kaplan
1
2
AI Is Watching What We Do, and That Is Good News
Advanced AI learns right from wrong from human behavior. The record, unlike the news, is mostly good.
Oct 2
•
Dr. Craig A. Kaplan
1
2
You Don’t Need to Keep Up With SuperIntelligence
AI will think at the rim of the wheel. Our place is at the center, where the values that matter barely move.
Sep 30
•
Dr. Craig A. Kaplan
1
See all
SuperIntelligence
The fastest path to superintelligence is the safest path.
Subscribe
Recommendations
View all 19
The Control Layer
Amer Altaf
State of AI newsletter
Nathan Benaich
AI Safety Events & Training
AISafety.com
Astral Codex Ten
Scott Alexander
Interconnects AI
Nathan Lambert
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
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
