Confusion on OpenAI's relationship to the field of AI safety

It was my impression OpenAI is concerned about AI safety, since it's backed by and founded by Elon Musk and Sam Altman who have expressed their concern for AI risks, and when interviewed on the topic of OpenAI, Musk and Altman made clear they think OpenAI's work, mission, and policies will bring the world closer to AGI while also ensuring it's safe. Multiple times now people have told me OpenAI isn't working on or about AI safety. I think what they mean is "AI safety" is something Nate points out in this article:

http://futureoflife.org/2016/01/02/safety-engineering-target-selection-and-alignment-theory/ Safety engineering, target selection, and and alignment research are 3 types of technical work/research, and more broadly, strategy research, moral-theory/machine-ethics, and collaboration-building are part of AI safety as well. So, it seems when people like Rob Bensinger or Victoria Krakovna tell me OpenAI doesn't seem like it will be doing AI safety in the near future, they won't be focusing on any of these areas. It seems to me OpenAI is, among other things, fostering alignment and capabilities research for AI in general without an explicit focus on the safety aspect. The 'open-source' component of OpenAI seems to be an effort towards creating a groundwork for strategy research or collaboration-building. Perhaps OpenAI is assuming safety engineering is part and parcel of capabilities research, and or perhaps that OpenAI can, with its influence in AI research in general, nudge capabilities and alignment research in the direction of safety concerns as well. My model of OpenAI's reasoning is that if all their research is the main force spurring capabilities research, it being open-source for everyone will level the playing field, not allowing any one company or other entity get ahead of the game without being examined by others, safety-concerned or not, and thus safety research can be injected into capabilities research in a broad way. Meanwhile, it seems folks like Scott Alexander, Jim Babcock and others have put forward that this approach is insufficient to precipitate AI safety research so it isn't outpaced by the relevant capabilities research, as it doesn't need to be a malicious entity, or one making philosophical failures, which makes AI dangerous, but technical failures in implementing an AGI which would also make it dangerous.

A month ago I made a big list of who is working on AI safety, as far as I could tell, and I included organizations like Google DeepMind and OpenAI because they've expressed a strong concern for AI safety and are now very big players in the field of AI more generally. Now, I'm understanding what people mean when they say OpenAI may not have anything to do with AI safety, because they're greatly mistaken about what AI safety really requires. So, I can exclude them from future versions of my list, or mention them but include caveats when I turn it into a proper discursive article. However, that still leaves the problem that some of the biggest players in AI in general think what they're doing will help with AI safety but it may actually make the future of AI more dangerous, and other big players in the field like Google might not worry about safety in their own AI research because they feel organizations like DeepMind and OpenAI have them covered. If this is the case, then it seems the mistaken understanding of the technical nature of AI safety needs to become as diffuse as increased awareness of it. This is a problem which needs solving.

The Death of the Republican Party Has Been A Long Time Coming

Everyone in the media is talking about how the Republican Party is collapsing. I think it's just become undeniable the Republican Party cannot be what it once was, what it wants to be, and what it was hoping it still might be. I don't think the collapse of the Republican Party started in 2016, or in 2014 or 2012 during the elections during which the Tea Party swept into success. I think the Republican Party started collapsing after the United States started gradually losing faith in their competence after the second term of George W. Bush. I think W. was the worst president in modern American history, at least since Nixon. As a Canadian, I have a different perspective, so maybe most Americans would disagree with me, but I think W. was even worse than Nixon. I'd venture W. might be the worst president in the post-War era, and I only say that because the pre-War era was a sufficiently different time, and also because I don't know the history well enough, to feel comfortable making predictions which go before that. I don't think it would be too hard for someone who knew their American history better to convince me W. was the worst president of the last 100 years, and one of the worst of all time.

Sure, there are plenty of Americans who feel every politician and Obama as well are wiping their butts with the Constitution, but there is still enough of a coherent core left of centre that the Democratic Party hasn't been imploding for the last 8 years, and isn't imploding right now.

Thomas Jefferson once wrote: "The tree of liberty must be refreshed from time to time with the blood of patriots and tyrants." As a modernist who has observed the quality of life in the developed world achieve wonders we're all lucky to have, I'm much opposed to bloody revolutions which would threaten all that is stable in democracies with elections that still work. However, I am glad Donald Trump is causing a crisis and a revolution in the Republican Party that is causing the spirit of the old guard to stare into the face of its doom. To cleanse the establishment so seems important enough sometimes I think it's worth the trade of putting Trump in the White House if the Republican Party dies. Barring authoritarian nightmare scenarios, Trump will only be the crazy President of the United States for four years until he's booted out in 2020, and then the Republican Party will be dead. As long as Trump can't succeed in violating human rights left, right and centre, his winning the Republican nomination for the Presidency might be good insofar as this will destroy the Republican Party and it will prevent a dangerous fundamentalist like W. from returning to the White House for a very long time as well.

On 'rationalist', 'effective altruist', and Labels-As-Personal-Identities

In a conversation about other things, Julia Galef of the Center For Applied Rationality mentioned as an aside she doesn't identify as an 'EA'. Of course, this is presumed by all to be shorthand for 'effective altruist'. I've been reading for a while these undercurrents that the identity of 'effective altruist', as a noun, as a personal label anyone can freely choose for themselves rather than a category they may or may not match, waters down what 'EA' really means, or could mean, and is becoming problematic. Of course, there are those who would say building this into some sort of ingroup-y, tribe-like identity has always been a problem, perhaps since its conception. It's a lot of the same problem many express with just about anyone identifying with the term 'rationalist', and that profession being accepted as long as that person can send the right signals, i.e., a surface-level understanding of the relevant memes, to the rest of the self-identified 'rationalist' community.

I know Rob Bensinger has for a long time expressed a preference for people referring to themselves as 'aspiring rationalists', or 'aspiring effective altruists'. I think this won't work, as that's so long a phrase as to strike most as unwieldy in casual parlance. At best, I think people will shrug, assume others will know the 'aspiring' is implied and implicitly tacked onto the use of the terms 'EA', 'effective altruists', and/or 'rationalists'. Of course, others *won't* or *don't* assume that, and then somewhere in each of our minds we assume we're 'bona fide' EAs, or rationalists, that we're the real deal, whatever that is supposed to mean. I think this has already happened. I don't perceive a way to enforce this norm of thinking of ourselves, not only as individuals, but as a collective, as constantly aspiring to these ideals as an asymptote to be approached but never achieved, us being the limited and epistemically humble humans we are, unless someone like Will MacAskill and/or Eliezer Yudkowsky wrote repeated implied injunctions against anyone referring to themselves as anything but 'aspiring', when relevant.

Ryan Carey once expressed a sentiment that 'effective altruist' is something, as the movement grows, will become a shorthand for those who are involved with the global community, the 'intellectual and social movement', that call itself "effective altruism". He predicted the term 'effective altruist' will come to refer to people who identify with it, without being especially problematic. This will happen in the same way 'Democrat' or 'Republicans' refers to Americans who identify with particular political parties, without anyone assuming someone affiliated with one party or the other being for democracy and against republics, or visa-versa. I rule this prediction is correct, and has already come to pass. I thus recommend people stop making as big a deal about how the term 'effective altruist' is used. I'm not as enmeshed with the rationality community, but for policy on what to think of and how to use the label and personal identity of 'rationalist', I defer to Scott Alexander's essay 'Why I Am Not Rene Descartes'.

Most people probably haven't noticed this, but I have stopped tending to use the term 'effective altruist'. Sometimes, in discourse when everyone else is using it, I feel forced to use the shorthand 'EA', or 'EAs'. It's just convenient and I don't want to break my or anyone else's flow. However, I mentally think of this as meaning a 'community member'. That is, an 'EA' is, to me, someone who has chosen to in some capacity be involved with the body of other people known as 'effective altruism'. The extent to which one is an 'EA' is the extent to which one involves themselves in this community. A 'hardcore EA' is someone who has perhaps made their involvement in effective altruism their primary community, as opposed to some other social category, subculture, social movement, etc. The abbreviation composed of two letters, 'EA', implies this, without implying one is necessarily someone is particularly effective, altruistic, or effectively altruistic. Some people who identify as EAs will not be particularly effective or as altruistic as they ought to be, and some who explicitly eschew the label will match in their actions the ideals of effective altruism better than most. In this sense I pragmatically accept 'EA'-as-community-member, while banishing from my personal lexicon thinking of some people as truly or real 'effective altruists', and others not being so. There are just individual humans who take actions, and ever bigger groups of them, be them subcultures, companies, or nations, who coordinate some of their actions or beliefs towards common goals. When there is greater coordination among greater number of peoples, it's to varying extents useful and efficient to think of them as a unified, cohesive, atomic agent.

I know this puts me in a position which may strike most as odd. I'm just putting my thoughts down for reference. However, I hope people will remove from the marker in their brain labelled 'EA' or 'effective altruist' that there's a strong correlation or implication that anyone who uses that term to refer to themselves as automatically way more effective, more altruistic, more rational, or otherwise better than anyone else they meet in life. There may be a weak correlation there, but to the extent you interact with individuals, EA, rationalist, or otherwise, get to know them first.

Evan's Dictionary

Sometimes I make up words. I think making up words is a valid strategy, especially for writers discussing relatively novel ideas, or novel perspectives on old ideas, if the majority of humans who would end up reading or hearing them would agree the meaning assigned to the word intuitively works. Thus, here are the words which, while I don't claim exclusive title to adding them to the English lexicon, seem rarely used before I thought of them myself. Additions to this dictionary are ongoing for the indefinite future. Suggested additions from friends of useful words they've made up and/or think I should also use, especially as the Evantionary see below) gains clout, are welcome and encouraged.

Dankularity: noun. A memetic dankularity is a joke and possibly a real prediction I've posited of some given subculture becoming primarily dominated by dank memes and other dankness-related phenomena as opposed to the domination of all other factors combined. 'Dankularity' as a word and idea is inspired by the idea of a technological singularity, as opposed to 'singularity' as its used in physics or other scientific disciplines.

Endarkenment: noun. The opposite of 'enlightenment'. Verb form: (to) endarken. Adj/Adv. form: endarken

Endodus: noun. The opposite of 'exodus'.

Evantionary: proper noun. A term I myself (Evan) and others may use to refer to 'Evan's Dictionary' (this here document). A portmanteau of 'Evan's' and 'Dictionary'.