I think there are probably a lot of people who tried ChatGPT a little bit in the beginning and then bounced off, or read all the articles about how LLMs hallucinate all the time and reasonably figured they didn't want to use them. But AI chatbots have gotten a lot better. My definite favorite is Claude ( ). (That website offers a few different models; the "Sonnet" version is the best, though requires a paid subscription if you want to talk to it in any regularity.) Here are some of the ways I've used it recently:claude.ai

Have been grinding these types of exercises for two weeks now

(It makes metaphysical claims about "energy" but I think it's mundane psychological and physical processes instead and the thing works anyway)

This is the closest that I've gotten to having "pleasure on demand"
A gentle touch feels pleasant and it turns out that an imagined gentle touch is pleasant, too

As it starts becoming practiced enough that I can access some of it at will, the consequence is a feeling of relaxation and widespread positive feelings in my body

A trauma book I was reading had an interesting claim that indecision is often because the person looks for the approval of an internalized authority figure but is unable to predict what action they would approve of. The writer is a Jungian therapist, so he attributed it to looking for the approval of an internalized parent, but I think it can be broader.

(Part I.) Sometimes I find my thoughts looping through the same motions.

I'm thinking of where to go to eat. Different options pop into my mind, none of them perfect.

"Ugh, that place is expensive"
"Ugh, that place has no good vegetarian options"
"Ugh, that place is far away" After a while, I notice my mind having bounced back to the beginning of the list.

"Ugh, that place is expensive..."

1000 hours of formal recorded meditation since January 18, 2018.

Doesn't include: probably a similar amount of unrecorded semi-formal meditation, a hard to estimate but significant amount of "off-the-couch" practice, practice I did after 2009 before starting to use this app. (Note that this screenshot has been slightly edited, since for some reason the "average per day" number it actually shows me is twice what it should be; the correct amount is 33.1 minutes [I couldn't be bothered with editing that last digit].)

I was feeling rushed this morning. It wasn't that I had any real urgency, but I want to get a reasonable amount of work done today, and I'd been having a slow start for the day. Besides work things, there were also several personal things that I needed to get done, and I was feeling an acute ugh that argh I need to do that and I need to do this and why didn't I do anything yesterday and now I'm going to feel rushed for the rest of the week again.

Thread of my favorite game intro cinematics Battletech.

600 years of future history compressed into two minutes. Gives me cold shivers each time.

I was recently asked how literal I consider the Internal Family Systems model of your mind being divided into "parts" that are kinda like subpersonalities

Short answer: more than just metaphorical, but also not as literal as you might think from taking IFS books at face value I do think that there are literally neurological subroutines doing their own thing that one has to manage, but I don't think they're literally full-blown subminds, they're more like...

Holy -

I'm getting cold shivers reading this paper

They made ML models trained on different modalities but all taking language input/output, to reason together by using language as a common representation

This is getting so close to "parts" models of the mind it's scary

https://twitter.com/andyzengtweets/status/1512089759497269251

My internal linguistic representation while reading this paper:

"Oh fuck"

"Whoa"

"Oh my god"

I was working with my (Internal Family Systems/Somatic Experiencing) therapist yesterday, and we ended up working on the memory of an incident where a bully had stolen money from me as a child. We did some imaginary re-enactment around it, and at one point my therapist asked if I would like to imagine what it would have been like if I'd had some sort of a guardian or defender as a child, protecting me from similar bullies.

woah, hadn't done Ideal Parent Figure practice in a while, had felt somehow vaguely averse to the thought

but listened to a guided meditation now

stuff.kajsotala.fi/IPF/IPF%20for%… spent most of the meditation in an image of being a baby, somewhere between a newborn and maybe six months old, just laying naked on the chests of my ideal parents, feeling into relaxation with skin against skin

Thinking about how this also applies to self-leadership. You can try to coerce yourself into doing what "you" want, or you can show your parts a vision of what would be awesome and have them all strive for that. "Focus your attention on what you want" (cc @QiaochuYuan @visakanv ) At the same time, there's a known risk with "positive thinking" in that you might just imagine a good outcome without doing anything for it. Some psych literature recommends contrasting your vision with your current situation to create motivation to act.
woopmylife.org/en/home

A thing that meditation teacher Tucker Peck says is that meditation doesn't make any states of mind unavailable, but it adds new ones to what it's possible to experience. In other words, doing a lot of meditation probably doesn't mean you'll ever _stop_ getting emotionally triggered or feeling miserable, but it does mean that there will be _more_ states of mind where you'll feel peaceful or at least okay with feeling bad.

Recently realized that this is Very Much Not the norm in every country: a single 500 employee company got into a dispute with a union, so the unrelated Transport Workers’ Union did a sympathy strike and put all Finnish cargo shipping on hold for a full day www-hs-fi.translate.goog/politiikka/art… The offense of the company was that it wanted to negotiate directly with its employees to determine their pay and other benefits, as opposed to being bound by a collective agreement negotiated with the union that sets the same minimum conditions for all companies in the field.

After I got Anders to play with wombo, he got me to generate pictures of my papers so let's go

https://twitter.com/anderssandberg/status/1468029590350221317

(all of these papers can be found from kajsotala.fi/academic-paper… btw)

People following some meditative traditions sometimes talk about the ego/sense of self/whatever-you-call-it as something bad, that you're supposed to destroy/dissolve/kill/etc.. That's a bit of a bleak image, so let me instead share with you one that I got at this week's retreat: At the back of my mind, there's a small robot that happens to look almost exactly like Wall-E. It sees my mind as a big warehouse with lots of boxes that need to be arranged. It would like everything to be neat and tidy and nice.

There's an online demo for an AI system that tells you whether different activities are moral, but I might not trust all of its judgments quite yet...

https://twitter.com/liweijianglw/status/1449080394616242176

https://twitter.com/egregirls/status/1450185051283599363

To be human is to tell yourself the same stories over and over again. I would say "until they become identity", but of course the flow of stories does not stop at that, nor is it limited to the things we would conventionally think of as identity.

One might say that identities are stable islands of self-fulfilling prophecy.

I mentioned elsewhere that I think the "System 1/System 2" terminology, as popularized by Daniel Kahneman's "Thinking, Fast and Slow" is a concept that has done more harm than good. This is in the context of the rationalist community and people who are trying to debug their own biases in practical terms - it may still be a valuable concept if you are an academic doing bias research.

Had alcohol for the first time in a little while yesterday, and after having spent some time with "be aware of the space around you"-type practices recently, it was a weird experience. When the alcohol was acting to dull and collapse my field of awareness, it felt like the part of my mind that I've been training to maintain broad awareness was putting in an effort to counter those effects, with some success.

(Part of) my corner of Twitter seems to have pretty strong consensus you don't need to be able to justify your preferences to anyone

https://twitter.com/m_ashcroft/status/1437776659164323846