LESSWRONG
LW

318
Chris Lakin
1994Ω633135831
Message
Dialogue
Subscribe

Deep unlearning researcher

LocallyOptimal.blog

@ChrisChipMonk

Sequences

Posts

Sorted by New

Wikitag Contributions

Comments

Sorted by
Newest
Boundaries (membranes) for AI safety by Chipmonk
High-level actions don’t screen off intent
Chris Lakin2d54

Mindreading is ubiquitous

Reply
Philosophy of Therapy
Chris Lakin7d20

Random: Maybe current lesswrong readers would like this post? It's been 5 years, maybe there's a way to re-post it

Reply
Kaj's shortform feed
Chris Lakin1mo20

When did you last use it?

Reply
On Pessimization
Chris Lakin1mo20

Do you find these examples relevant? Examples of self-fulfilling prophecies in AI alignment

Reply
Four types of approaches for your emotional problems
Chris Lakin1mo20

Thanks for writing this!

Reply1
Emotions Make Sense
Chris Lakin1mo50

I’ve only encountered a handful out of a few hundred teenagers and adults who really had a deep sense of what it means for emotions to “make sense.”

How would you make sense of the emotion of doubting the value of other emotions?

Reply
Emotions Make Sense
Chris Lakin1mo40

This is the point in the class where I ask participants to pick an emotion, any emotion, that they feel is bad, or wish they didn’t have, or think the world would be better off without, and spend 3 minutes trying to generate the reason it exists, and might be worth having after all.

 

For anyone who wants to check how they did on this, you can copy this whole post into AI (I recommend o3 or Claude 4 Opus with extended reasoning on) and ask it to be the author of the post without immediately giving answers

Reply
Emotions Make Sense
Chris Lakin1mo71

Thank you for writing this! I will be linking to this

Reply
Launching Forecast, a community for crowdsourced predictions from Facebook
Chris Lakin2mo20

What came of this?

Reply
A night-watchman ASI as a first step toward a great future
Chris Lakin2mo40

Related: What does davidad want from «boundaries»? 

(also the broader work on boundaries for formalizing safety/autonomy, also the deontic sufficiency hypothesis)

Reply
Load More
188“Flaky breakthroughs” pervade inner work — but almost no one tracks them
3mo
39
141Social Anxiety Isn’t About Being Liked
4mo
21
18Alignment first, intelligence later
Ω
5mo
Ω
5
61Softmax, Emmett Shear's new AI startup focused on "Organic Alignment"
Ω
6mo
Ω
2
20Examples of self-fulfilling prophecies in AI alignment?
Q
6mo
Q
7
54Just one more exposure bro
9mo
6
41Locally optimal psychology
10mo
7
8What is autonomy? Why boundaries are necessary.
11mo
1
63Pay-on-results anxiety coaching: first success
1y
8
37I didn't have to avoid you; I was just insecure
1y
7
Load More