17,129 karma · joined March 19, 2019
Interests in self organizing systems and cooperative economic structures.
For inquiry or long term discussion, please feel free to reach out to me: s1artibartfast.hn.correspondence at gmail
Some SBF Quotes
> "This dumb shit I said—it's not true. Not really. Every soul-searching, virtue-signaling thing I said was just part of the game."
> "Fuck regulators. They make everything worse. They don't protect customers at all."
The concept is called "patient phlithoropy", apparently about 1% of EA donations go to such funds.
There is no pope of EA that sets doctrine, so anyone can do and say what they want.
I have heard a similar argument that you dont need to sell everything you own and donate it all, because a steady X% per year is better than a lump sum and then homelessness. Or that it makes sense to spend on a college degree so you can earn and donate more later. Both make sense to me.
That said, basically every philanthropic org makes some degegree of "wait until later to be more effective". The gates foundation or churches dont try to spend every penny they have in the current year.
It doesn't seem like most people are taking the agnostic "wait and see" approach when it comes to judging effective altruists (or vegans).
To accept that someone else might be right on a moral issue also means accepting that you might be morally wrong.
I don't send 10% of my income to charity or eat vegan, but it does seem to me that people feel threatened when they see others doing this
The same is true for science and Engineering.
What matters is if you have a method of sorting through the repetitive trash
According to AI, the biggest EA charity is GiveWell, which spends about 600 million per year. 95% of the money on low-cost high-impact global health programs like mosquito Nets and vitamins for children.
There is also open philanthropy, which gives about 500 million per year With a wider perspective. Roughly 50, 40 and 10% go to Global Health, Existential Risks, and then Animal Welfare in that order.
That said, I'm not really trying to argue the case. If the simple reason is that people have characterized EA based on SBF, and that's how they think about these things, then that is a complete answer.
Is that the standard you hold everything to?
At best they are just people trying to donate money to effective charities like mosquito nets in Africa. At worst, some might be wasting their money on animal welfare or AI safety. This doesn't seem like enough to fuel the kind of backlash I see.
The exact formula will change depending on the scenario. Maybe it is a prompt email, maybe it is getting a beer and giving someone a pieceof your mind. Most important, it is knowing when to do which one.
AI creates novel discovery> incorporates this information > makes new discovery
This is how it works for humans too.
About $310B+ has been invested in OpenAI and Anthropic represents their total venture funding over their entire corporate histories.
The majority of doctor's offices and hospitals are already nonprofit. Many of them are state-run.
By contrast, the U.S. spends $14 Billion to $15 Billion every single day on healthcare, around 5 trillion per year.
That's why you have the family of the deceased reading tearful letters, crayon drawings from children missing their parents, and other things of that nature introduced during pre sentencing impact statements.
Family can read fictional poems, personal thoughts, videos or photo montages. They can put up a picture of the deceased, and make statements like if "Bob was here, he would say x, y, and z"
The point is to influence the sentence.
Provided that the family was very clear this was not historic video, and I think it was, then one can see how this isn't obviously out of bounds
It is a neutral role with broad applicability. They could equally be trained for other styles. A model trained to talk as a pirate would get old, especially after it told you to bugger off when you asked for a cupcake recipe. A model trained to write like Hemingway would be frustrating when you want directions.
What would be better? who should it write like?
There is no technical challenge finetuning models to write in a specific style that is indistinguishable from a human
You can do a lot without designated and built roads, as buring man up to that point obviously demonstrated. If it wasn't constantly growing, they probably could have gone several years without another death
Also drive on beaches would be a good example.