293 karmaJoined


Sorted by New


I think this article paints a fairly misleading picture, in a way that's difficult for me to not construe as deliberate. 

It doesn't provide dates for most of the incidents it describes, despite that many of them happened many years ago, and thereby seems to imply that all the bad stuff brought up is ongoing.  To my knowledge, no MIRI researcher has had a psychotic break in ~a decade. Brent Dill is banned from entering the group house I live in. I was told by a friend that Michael Vassar (the person who followed Sonia Joseph home and slept on her floor despite that it made her uncomfortable, also an alleged perpetrator of sexual assault) is barred from Slate Star Codex meetups. 

The article strongly reads to me as if it's saying that these things aren't the case, that the various transgressors didn't face any repercussions and remained esteemed members of the community.

 Obviously it's bad that people were assaulted, harassed, and abused at all, regardless of how long ago it happened. It's probably good for people to know that these things happened. But the article seems to assume that all these things are still happening, and it seems to be drawing conclusions on the basis of those assumptions, e.g. that misogyny is currently ubiquitous in the community, that AI alignment is a toxic field to work in, that people are regularly having psychotic breaks, etc. 

For what it's worth, I think (as someone who has firsthand experience with EA and rationality in the Bay Area) that these things are untrue. But even if they were true, the author has not demonstrated that they possess the evidence to draw these conclusions, and yet they are nonetheless trying to convince the reader of them.

I would be interested in a Zoom discussion.

I like this pitch outline; it's straightforward, intuitive, and does a good job of explaining the core ideas. If this were to actually be delivered as a pitch I would suggest putting more focus on cognitive biases that lead to inaction (e.g. the human tendency to disbelieve that interesting/unusual/terrible things will happen in one's own lifetime, or the implicit self-concept of not being the "sort of person" who does important/impactful things in the world). These are the sorts of things that people don't bring up because they're unconscious beliefs, but they're pretty influential assumptions and I think it's good to address them. 

For instance, it took me some doing to acquire the self-awareness to move past those assumptions and decide to go into x-risk even though I had known for quite a while on an intellectual level that x-risk existed. It required the same sort of introspection that it did for me to, when I was offered a PETA brochure, notice my instinctive negative reaction ("ew, PETA, what a bunch of obnoxious and sanctimonious assholes"), realize that that was a poor basis for rejecting all of their ideas, and then sit down and actually consider their arguments. I think that it is uncommon even for bright and motivated people to naturally develop that capacity, but perhaps with some prompting they can be helped along.

I would also be interested. I haven't read very many LW articles and I don't know very many people here, so it would be nice to get caught up in the context of a discussion.

Great post! As a tofu enthusiast myself this is pretty exciting on a personal level. That being said, you did ask for concerns/criticism, so:

If our goal is to create new foods, the best people to do so are restaurant chefs. 

I'm not sure that this is true. Americans are well-known for eating out often, but what's the actual frequency? I wasn't able to find a reliable survey with some cursory googling, but from personal experience and what I did see (for instance, this) it seems like getting a majority of meals from eating out is rare. In this case, perhaps it would make more sense to focus on working with amateur cooks to produce recipe blogs and cookbooks? There are a lot more amateur cooks than restaurant chefs out there, and many are quite interested in experimenting with new ingredients/creating new recipes.

The types of food usually produced in a restaurant setting are also pretty different from the ones people would want to cook for themselves; for instance, in the latter case, simplicity/ease of cooking and low cooking times are far more important.  So it's not necessarily true that restaurant chefs developing new dishes would easily translate to a greater variety of dishes people would be able to cook in their own kitchens. If people prepare most meals at home, it seems like it might be more impactful to focus efforts in that area.

From a first principles perspective, given rates of flexitarianism and the fact that tofu is by far the most culinarily-versatile of global plant-based proteins, consumption could easily be 4-10x higher. 

You justify this by saying that there are some foods Americans eat quite often, but all of those foods are ones easily prepared in meals at home. It seems reasonable to presume that the invention of new and tasty tofu dishes will incentivize people to eat them when they eat out, but it's pretty doubtful that the frequency with which people eat out will significantly increase.

Do you know of potentially allies I should reach out to? I would really appreciate any warm intros.

I am part of a vegan/vegetarian club at my university, and we keep in touch with various local plant-based food businesses; many of them have track records of creating innovative new dishes, and would likely be happy to experiment with rare tofus. These are generally small-scale operations, so I'm not sure if that's what you're looking for, but if you're interested in collaborating with them I can reach out. My club would also likely be willing to promote the project on social media (and, more generally, perhaps recruiting student groups with an interest in plant-based foods could be valuable to your social media campaign).

Thanks for pointing me in the direction of these resources! I just signed up for the Alignment Newsletter.

As a note, the AN link you posted actually doesn't work, though I was able to figure it out with my fabulous tech skills (the URL has a ] at the end). Here's a working link, for posterity.

Hello, my name is [name redacted] and I'm a new member here. Or, more accurately, I've been here for a little over a month but I haven't introduced myself yet because writing on a public forum is mildly anxiety-inducing for me. However, I have lately been attempting to become better at things I'm bad at by doing things that make me uncomfortable; this has included applying for an internship at Redwood Research despite not feeling terribly qualified, spending an afternoon using food to entice college students into discussing the merits of not eating meat despite being socially anxious, and now making an introduction post despite my instinctive aversion to doing so.

I am an undergraduate at a non-prestigious state university in the US studying computer science and math. I'm also working part-time as a research assistant to a professor working on a reinforcement learning project. The latter is a fairly recent development, and I find it a somewhat astounding one; about a year ago I was expecting to spend my time at university using scholarship funds to eke out a meagre existence, whereas now I will be making enough money that I could eat out 2-3 times per day (not that this seems like the best course of action). I have briefly considered donating some of my newfound income, but this seems rather unwise considering I am still not exactly financially secure.

Approximately a year ago I read Nick Bostrom's book Superintelligence, which greatly changed how I viewed the world. I would like to work in AI alignment and am trying to take steps to get there, but I'm not sure if I'm going about it in the right way. If anybody is willing to offer advice that would be appreciated. I am also looking for friends who are similarly interested in positive impact, as I am currently rather lacking in that department.

I am not the best at writing in these sorts of contexts, so I will cap off my introduction by shamelessly plagiarizing myself from when I've written in other contexts.

Here's something: "When I was a small child I was very interested in the notion of being immortal. One of my early unpleasant memories was crying in my bed because one day my cat would die, and then (hopefully at the very least separated by a good number of years), I would die as well. This was quite horrifying to my seven-year-old self, and I liked to shield myself from that by imagining things like being a vampire, or an immortal magical cat. Sometimes I did both simultaneously; memorably, at some point I came up with a vampire cat that, rather than drinking blood, imbibed the color purple for sustenance. This served the double purpose of reducing suffering for other cats, and of ridding the world of my least favorite color. The fact that it would be swimming in my innards did not, apparently, bother my younger self."

And here's another thing: "Something else I find amusing is doing things in the vein of: Coming up with the string, "sad bad mad lad fad," and then trying to picture what that would look like. In this case, I envisioned a phenomenon wherein male youth become violently disillusioned with society and take to the streets to cause mayhem and destruction. It's only a fad, though. This year the lads are using homemade explosives to destroy public infrastructure in a manifestation of their rage at the unfulfilled promises of the American Dream, Fight Club style, but next year? Who knows? Maybe there'll be an uptick in anime cosplay?"

I feel sufficiently introduced now, and thus I will take my leave. I would like to reiterate that I am looking for friends, so if anyone reading this finds me to be plausibly friend-worthy I would appreciate it if you sent me a private message.