3 ms·

Obtaining the correct categorization of something is not social engineering, it's returning a FACTUAL answer. I didn't ask the AI "are men more likely to be doc

Obtaining the correct categorization of something is not social engineering, it's returning a FACTUAL answer. I didn't ask the AI "are men more likely to be doctors" I asked "what is a doctor" Do you understand the difference? I don't expect it to tell me "Doctors are male individuals who...." No where in my original post did I say the AI has to lie about gender ratios, I would expect any AI to tell me also, the socio-historical-cultural perspective of the medical profession, and how it has changed over time. I am simply not interested in a neat little, static, steady state view of something that conveniently leaves out context. Do conservatives want to exist in a bubble where no criticism of the status quo is seen? If a machine learning algorithm learned a bad definition of conservative, that cast them as crypto-racists, you'd want it corrected wouldn't you? Even if a reading of conservative news sites comment forums gives you exactly that impression is likely true? > "Quite why this outcome is better than the existing status quo is never explained with any rigour - it's ideology." Right, progressives are the only ones with ideology, conservative positions are arrived at by cold, hard logic? Tell me why an African American slave boy, growing up in the 1800s, who learns the existing status quo that black people are property, and white people or not, is a worse outcome? Clearly, it's a better outcome for white people of the era, so any explanation for why it might be preferable to alter it has to argue that the status quo wasn't good for black people, or that it wasn't good for white people for some presumably economic reason. Maybe just maybe, the status quo isn't good for women? Maybe it would be good to ask them if it's ideology, or if changing cultural attitudes about what women are allowed to do, and capable of doing over the last 200 years has been a positive change in the status quo for them? BTW, Conservatives don't reject social engineering, they just reject social engineering they disapprove of. Social conservatives around the world, in concert with religion, have sought to engineer human behavior with appeal to cultural and religious piety, and in many cases, winning laws that enforce such behavior. We've had anti-sodomy laws, anti-booze laws, anti-miscegenation laws, all of them enacted by social conservatism. And what do you call religious proselytizing, if not social engineering? Trying to spread memes and infect and convert more people into a new way of thinking. Sometimes I feel that conservatives are against secular, humanist 'social engineering' the way Scientologists are against Psychotherapy, because it's competing in the same meme space. Society is a dynamic equilibrium. It is constantly evolving, sometimes it evolves purely organically, in a spontaneous order, and sometimes there are clusters and movements that boil over, and change arrives by deliberate persuasion. Google is a global, transnational company, that serves the entire world, 7+ billion people. It needs to reflect that diversity. And like I said, it absolutely cannot have AI that does stupid stuff like learn that black faces are gorillas.

repolfx 8y ago
I didn't ask the AI "are men more likely to be doctors" I asked "what is a doctor" I'm afraid you're obfuscating. The most important paper in question is this one: http://papers.nips.cc/paper/6228-man-is-to-computer-programmer-as-woman-is-to-homemaker-debiasing-word-embeddings.pdf http://papers.nips.cc/paper/6228-man-is-to-computer-programm... It documents the researcher's "discovery" that word embeddings trained on all available text learn that most doctors are male and most nurses are female, along with many other relationships, like volleyball / football being a female / male analogy. Word embeddings can't answer a question like "what is a doctor", even people would struggle to give a good answer to such a vague question. So they asked it specific questions about gender, namely, "if you ask for a gender relationship starting from doctor what do you get" (answer: nurse). And then they decided this was bias, and wrong, and should be edited out of the model. So yes - if you asked such a model "are men more likely to be doctors" it would answer "no" although the correct answer is "yes". I wonder if that changes your views? Right, progressives are the only ones with ideology, conservative positions are arrived at by cold, hard logic? Both positions arrive at their conclusions via logic, but working forward from different assumptions and premises. The difference is progressives are much more likely to try and impose change on the world top down, whereas conservatives are much more likely to leave the world be (note: attempts to remove top-down imposition of progressive views is often cast by progressives as equally equivalent "change", but it's not, likewise, not attempting to change old traditions is often cast as inherent support for them rather than a general aversion to imposing top-down change). Your post is a great example of the dangers of this attitude: Maybe just maybe, the status quo isn't good for women? I happen to think the status quo, where women are routinely handed jobs and money simply for being women (e.g. the whole 'women in tech' movement), is excellent for women! But our views on this are both irrelevant because it's not what the argument is about. The point is, do we want to build AIs that understand the world as it is today or which have been given an understanding of the world as Googlers believe it should be. You are very clearly arguing for the latter, here, to the extent that earlier in your post, you argued a good answer to the question "what is a doctor" would actually involve a lecture on the "socio-historical-cultural perspective". Who says users of AI give a crap about any of that? Maybe they just want to get a definition of the word doctor, without some AI trying to change their kids ideas about what job they'd like to do along the way? A Google AI that is constantly engaging in proxy social advocacy with me would be annoying as hell. A Google AI that doesn't but has a warped and distorted view of reality because its creators feel they're towering moral actors with a mandate to change the world on behalf of billions of people they never met? Even worse! I'd rather it was at least open about it. Look at it this way. If Google AI was constantly subtly suggesting that all progressives were naive, hated America, that government intervention always failed and markets were the best way to do things, <insert random political stereotype here>, you wouldn't be very happy about it, would you? Especially not if the AI hadn't actually learned such things but such beliefs had been added later by libertarian programmers convinced that they'd make the world a better place by doing so. We've had anti-sodomy laws, anti-booze laws, anti-miscegenation laws, all of them enacted by social conservatism I suspect there's a slight terminological difference here between libertarianism and conservatism (I've been meaning primarily the former). But regardless, are you sure you aren't assuming that? Let's take anti-booze laws. Prohibition in the USA was a bipartisan issue at the time and both Democrats and Republicans voted in favour of it, in fact more Democrats voted in favour than Republicans did. The constitutional amendment that gave women the right to vote was introduced by a Republican. It took a long time to get passed (decades), but this was partly because - just like the UK - the woman's suffrage movement hurt its popularity through aggressive tactics and association with unrelated social policies, in particular with pacifism and refusal to join World War 1. And as for religion, I'd note that the USA is probably the most conservative (or libertarian) country in the world and it's also famous for separation of church and state.

repolfx 8y ago

I didn't ask the AI "are men more likely to be doctors" I asked "what is a doctor" I'm afraid you're obfuscating. The most important paper in question is this one: http://papers.nips.cc/paper/6228-man-is-to-computer-programmer-as-woman-is-to-homemaker-debiasing-word-embeddings.pdf http://papers.nips.cc/paper/6228-man-is-to-computer-programm... It documents the researcher's "discovery" that word embeddings trained on all available text learn that most doctors are male and most nurses are female, along with many other relationships, like volleyball / football being a female / male analogy. Word embeddings can't answer a question like "what is a doctor", even people would struggle to give a good answer to such a vague question. So they asked it specific questions about gender, namely, "if you ask for a gender relationship starting from doctor what do you get" (answer: nurse). And then they decided this was bias, and wrong, and should be edited out of the model. So yes - if you asked such a model "are men more likely to be doctors" it would answer "no" although the correct answer is "yes". I wonder if that changes your views? Right, progressives are the only ones with ideology, conservative positions are arrived at by cold, hard logic? Both positions arrive at their conclusions via logic, but working forward from different assumptions and premises. The difference is progressives are much more likely to try and impose change on the world top down, whereas conservatives are much more likely to leave the world be (note: attempts to remove top-down imposition of progressive views is often cast by progressives as equally equivalent "change", but it's not, likewise, not attempting to change old traditions is often cast as inherent support for them rather than a general aversion to imposing top-down change). Your post is a great example of the dangers of this attitude: Maybe just maybe, the status quo isn't good for women? I happen to think the status quo, where women are routinely handed jobs and money simply for being women (e.g. the whole 'women in tech' movement), is excellent for women! But our views on this are both irrelevant because it's not what the argument is about. The point is, do we want to build AIs that understand the world as it is today or which have been given an understanding of the world as Googlers believe it should be. You are very clearly arguing for the latter, here, to the extent that earlier in your post, you argued a good answer to the question "what is a doctor" would actually involve a lecture on the "socio-historical-cultural perspective". Who says users of AI give a crap about any of that? Maybe they just want to get a definition of the word doctor, without some AI trying to change their kids ideas about what job they'd like to do along the way? A Google AI that is constantly engaging in proxy social advocacy with me would be annoying as hell. A Google AI that doesn't but has a warped and distorted view of reality because its creators feel they're towering moral actors with a mandate to change the world on behalf of billions of people they never met? Even worse! I'd rather it was at least open about it. Look at it this way. If Google AI was constantly subtly suggesting that all progressives were naive, hated America, that government intervention always failed and markets were the best way to do things, <insert random political stereotype here>, you wouldn't be very happy about it, would you? Especially not if the AI hadn't actually learned such things but such beliefs had been added later by libertarian programmers convinced that they'd make the world a better place by doing so. We've had anti-sodomy laws, anti-booze laws, anti-miscegenation laws, all of them enacted by social conservatism I suspect there's a slight terminological difference here between libertarianism and conservatism (I've been meaning primarily the former). But regardless, are you sure you aren't assuming that? Let's take anti-booze laws. Prohibition in the USA was a bipartisan issue at the time and both Democrats and Republicans voted in favour of it, in fact more Democrats voted in favour than Republicans did. The constitutional amendment that gave women the right to vote was introduced by a Republican. It took a long time to get passed (decades), but this was partly because - just like the UK - the woman's suffrage movement hurt its popularity through aggressive tactics and association with unrelated social policies, in particular with pacifism and refusal to join World War 1. And as for religion, I'd note that the USA is probably the most conservative (or libertarian) country in the world and it's also famous for separation of church and state.