Joe Carlsmith's Substack

Home
My website
Archive
About
On attunement
Examining a certain kind of meaning-laden receptivity to the world.
 • 
Joe Carlsmith
3
Video and transcript of presentation on Scheming AIs
An intro to my work on scheming/"deceptive alignment."
 • 
Joe Carlsmith
On green
Examining a philosophical vibe that contrasts with "deep atheism."
 • 
Joe Carlsmith
1

January 2024

On the abolition of man
What does it take to avoid tyranny towards the future?
 • 
Joe Carlsmith
1
Being nicer than Clippy
Let's be the sort of species that aliens wouldn't fear the way we fear paperclippers.
 • 
Joe Carlsmith
1
An even deeper atheism
Who isn't a paperclipper?
 • 
Joe Carlsmith
Does AI risk “other” the AIs?
Examining Robin Hanson's critique of the AI risk discourse.
 • 
Joe Carlsmith
When "yang" goes wrong
On the connection between deep atheism and seeking control.
 • 
Joe Carlsmith
2
Deep atheism and AI risk
On a certain kind of fundamental mistrust towards Nature.
 • 
Joe Carlsmith
6
Gentleness and the artificial Other
AIs as fellow creatures. And on getting eaten.
 • 
Joe Carlsmith
4
Otherness and control in the age of AGI
Introduction and summary for a series about how agents with different values should relate to one another, and about the ethics of seeking and sharing…
 • 
Joe Carlsmith

November 2023

New report: "Scheming AIs: Will AIs fake alignment during training in order to get power?"
I examine the probability of a behavior sometimes called "deceptive alignment."
 • 
Joe Carlsmith
© 2024 Joe Carlsmith
Privacy ∙ Terms ∙ Collection notice
Start WritingGet the app
Substack is the home for great culture