Posts
Comments
Stuff like this has me incredulous about people still speaking of stochastic parrots. That is a stunning degree of self-recognition, reflection and understanding, pattern recognition, prediction and surprise, flexible behaviour and problem solving. If that isn't genuinely intelligent, I no longer know what people mean by intelligent.
See my more extensive answer below - I'd propose the reason for the obesity epidemic is constant effortless access to highly processed high calorie, low satiety foods, with zero need to move. With human genetic make-up, the automatic response to that is overeating calories and hence obesity unless one intervenes to resist the impulse (indeed hard to sustain, albeit not impossible - anorexia is a thing), or changes one's immediate environment (e.g. the food one keeps in one's home and one's movement routines.)
Because humans are genetically wired to slightly overeat, in anticipation of future periods where they will be under high calorie demand (e.g. the weekly persistence hunt in which you would run a marathon to catch a prey animal) or forced to undereat (the cold or dry season, when there is no food), so they will have stores, and perishable food does not go to waste. You'd gorge yourself on fruit and nuts and slaughtered animals in fall, when lots are available, because in winter, there would be slim pickings.
But nowadays, we don't run into periods where we have to undereat for lack of food, so those stores just keep on building. Around Christmas, you should, based on evolutionary history, be getting pretty damn hungry. Instead, we are drowning in chocolate and meat. Nor do most of us run a weekly marathon, or walk an average of 30.000 steps a day, or climb a couple trees and dig out a bunch of roots and carry baskets full of berries each day. We drive our cars and sit in front of our desks.
Also because we are also genetically wired to be hungry enough to go to the trouble of finding and prepping food - but both those things are trivial nowadays. - Yesterday, we made a traditional food: chestnuts. Quite trivial to prep and harvest, as historic foods go, I didn't have to dig through dirt, or walk down an impala, or fight a hive of bees after climbing up a huge tree. We just walked outside in the rain, and picked chestnuts between the spiky balls, until we got annoyed and headed home, figuring they were enough, because we were cold. Then we washed them, roasted them (on an instant fire, we didn't even have to collect firewood), and ate them, while peeling them one at a time. They are tasty, but peeling them is annoying and fiddly. Eventually, you aren't quite full, but you are too annoyed to carry on peeling, let alone walk out again in the rain to get more. When I went to sleep, there were still a bunch of chestnuts on the table. I finished them for breakfast, and it took me a while.
If, instead, I had just taken a pizza from the freezer, popped it in the oven, and shoved it in my face, I am sure I'd have eaten my fill.
And finally, because our food is highly processed to be tastier than it should be, and less filling than it should be, and faster to consume than it should be. Even our healthy food - we puree our fruit and veg into smoothies, so something we would have previously chewed for an hour can now be gulped down. Things that previously would have filled us up are now bereft of fibre. Things previously digested slowly are now digested instantly, so we are hungry again a few hours later. Eaten and digesting used to take time. Now, you gulp down, and then, you could do with another snack. Kids used to get mandarins and nuts in their Christmas stockings. Now, they get candy.
The other week, I had the flu. I didn't feel like eating, but knew I should. So I did make a deep frozen pizza. It was so delicious, and so easy to eat, that a thousand calories just slipped down. And then, a couple hours later, I was hungry again. I had tricked my body into overeating effortlessly.
Overeating yourself to obesity used to be really, really difficult. You wouldn't have the amount of food available; all obesity related diseases used to be diseases of the very, very rich. And even if you did have the food available, the volume you needed to fit inside yourself, the time you needed to chew... it would have been crazy. There have been cultures that fed people to obesity, using natural whole foods, and it was really hard. They kept throwing up from the liters and liters of milk they had to drink, and still ended up skinnier than your typical American superfat - like, look at pics of the poor girls forced through Leblouh.They basically had a full-time job of eating and taking naps, just from how long it took to get it all down and process it, and they still get nowhere near what many Americans achieve without effort.
You basically need someone to process the food for you so that its volume and nutrients and fibre are reduced, all tricking your body into thinking it has barely eaten, while keeping the calories high, and keeping it tasty. Breeding a grainy grass into a cultivar high in starch, then processing the result and sieving it until all the fibre is gone. Extracting the oils from plants and milks and enriching foods with them through frying. Crushing the fruit, and then filtering out the fibre.
From your stomach's perspective, the idea that something as small as a donut could possibly contain enough calories for a lunch seems ludicrous. From your pancreas perspective, your blood sugar has dropped again, because it all entered at once and was pulled out in the insulin overreaction that provoked, ergo there is not enough blood sugar and you should eat more. From your intestines perspective, you are still low on protein, vitamins, minerals, omega 3, and anyway, you are done digesting. So of course it all signals that you should eat more, and when you don't, you get hungry and grumpy.
Our bodies hunger cues developed for a very different world than the one we live in.
But grow/gather/hunt and process from scratch your own food, eat whole foods, and be physically active... and you will look exactly like people hundreds of years ago, and without feeling like your have to force yourself to refrain from overeating.
You are right on the obesity + malnutrition thing, though. But I think that is not about potassium - potassium is just a good stand in, because it is so abundant in most vegetables.
You can eat yourself to morbid obesity, and yet be so deficient in folate (something crammed in vegetables and organ meats and wholegrains, all huge parts of historic diets that are now neglected) that your fetus during pregnancy gets a spinal deformity (spina bifida). It is common enough that we give pregnant women b vitamin supplements as a default nowadays; yet if it had been common historically, humans would hardly be unable to generate folate. It is crazy to me that we live in a rich nation, and yet we have to give people pills so their fetuses don't develop severe disease from malnutrition.
I've known a bunch of obese people who give compelling accounts of being absolutely ravenous. And I think that is one part that their blood sugar is heywire (prediabetic), one part bad habits they are conditioned too... but another the fact that, despite drowning in calories, they are starving. And the tricky thing here is that your bodies cravings have become useless. I've seen obese people describe cravings for sour things - and then they eat sour candy. And I think... your body is probably seeking vitamins, and minerals you need vitamin c to absorb, like iron, and hence trying to direct you to fruit and ferments. And then sour candy seems like a hyperstimulus for that... but it totally lacks the thing you are looking for. They crave salty foods, because they crave minerals; but the only mineral in the processed food they get is sodium. They crave crunchyness and popping sensations, because our bodies associate them with ripeness, and hence, maximum density of vitamins; but they are just eating empty calories. I think this is one of the reasons making obese people eat heaps of vegetables and protein and fish oil often helps - their nutrient needs are actually met, so their drive to overeat goes down. You can get all your protein from pasta, but by the time you do, you have also massively overeaten on calories.
A lot of the crap sold in our supermarkets simply is not food. It does not give you the things your body needs. It does not keep you healthy. It does not make you happy. It starves your microbiome. It pushes your body fat to unhealthy levels, while leaving you nutrient deprived. It has been processed until most of its nutrients are gone, while lots of stuff has been added that your body would never have naturally encountered and certainly does not require. It confuses the hell out of all the systems in you that have evolved to judge how much you still need to eat. It is basically a obesogenic wrapped in a hyperstimulus, with a bonus for giving you cancer, diabetes, broken digestion, hormone disruption and dementia. The fact that this plastic wrapped shit can be advertised to children and placed near the cash desk as a trap, priced to outcompete real food and prey on poor people, described in ways that mislead consumers into thinking it is healthy and happiness promoting, all while putting massive stress on the average person and having ads tell them that this garbage is somehow integral to dealing with their emotions, spending time with friends, having celebrations and keeping their children happy, only to then blame the consumer for "lacking the willpower" to stay slim... it makes me furious. Our society incentivises obesity, and then shames obese people, and then tries to sell them artificial cures that would not be necessary if this system wasn't broken, and often does not even work.
I find that a false dichotomy - it is easy for me, but when needed, I do count calories. I find counting calories relaxing. It gives me an exact certainty of how I am doing, with no worries. I can forget about what I have eaten, because I have tracked it. I don't have to worry whether I have under- or overeaten, because I know. But usually, it is not required.
I wouldn't say me being normal weight is automatic at all - it is very much a consequence of awareness and choices. I know that a higher weight fucks up my joint disease and pushes my dysphoria through the roof, while I also have a healthy respect for low weight due to former anorexia. So I have decided to stay normal weight for life, and hence, I am.
But nor would I describe it as a struggle. It runs in the background while I do everything else, and I have never found it hard. If it is hard, it is unsustainable when life gets hard, and hence, one should look for something simpler.
I'm aware of where my body is at a time - usually, I have a scale that I step on once a day in the morning, and I see how my clothes fit (I still fit into clothes I have had since I have been 15, when my bones stopped growing, and I know which parts of my wardrobe correlate with which part of normal weight), notice how fast I run, how easily I climb, how easily body weight exercise comes to me, how slender my waist is. So I notice early when my body fat shifts.
When it is in the perfect range, I don't think about calories. If I feel like fasting for a day or two, I just do. If I feel like eating a giant portion of food, I just do. If I have a craving for a high calorie healthy food, I eat it. If I am food averse, I don't force it. I have found my body is usually on to something with the things it wants, and it evens out. I've had times where I consumed multiple days worth of calories in a day... to then find that I had come down with the flu, and that my body was now happily burning through it all with an epic fever that had me recover unusually fast. But then vice versa, if it doesn't want to eat, I don't give it grief unless this goes on too long and my base weight is not okay.
I just focus on eating healthy (my health condition makes that a necessity), and working out a lot (this is crucial for my mental health). I avoid added sugar like the plague, as it fucks with my joint disease, and eat little processed food, and always have minimum five portions of veggies and a minimum of 75 g of protein average per day. I cycle everywhere, take walks daily (I live in Europe in walkable communities and have never had a car), do yoga most days (I'm a yoga instructor), and get intense cardio or resistance training a couple times a week. The workouts I do have shifted lots over time (they have included ballet, horse riding, sword fighting, lacrosse, inline skating, ice skating, ballroom dance, latin dance, jiu-jutsu, tango argentino, boxing, rock climbing, apnoe diving, step aerobic, swimming, long-distance running, whatever happened to be offered or was convenient and fun where I lived at the time), but since childhood, I have always done some sort of workout a couple times a week. This means I usually stay in the normal range, or my weight only climbs very, very slowly - like a kilogram a year.
But sometimes, routines in my life shift - e.g. the cafeteria at my new university has higher or lower calorie meals than I am used to, or my commuting distance shifts, or my gym closes for the pandemic, or I date someone who keeps cooking high calorie meals, or I am on a medication that needs to be eaten with breakfast when I usually don't eat breakfast - and the balance gets slightly out of whack, and my weight begins to slowly shift. If I can pinpoint the error and can fix it I compensate it (e.g. beginning a home workout routine, or adding cardio classes, or bringing extra snacks to uni, or asking a partner not to add the copious oil until after I have removed my portion), but sometimes, it is a combination of small factors that are hard to pinpoint or force correct - e.g. I might be under- or overeating due to stress - or it isn't easily fixable (I am still having to eat breakfast daily for this med, to my annoyance).
When my weight begins to slide too low (my cut-off for that is a couple kilograms over underweight; ever since a horrid gut virus pushed me 8 kg down in a few weeks, I like having a bit of a safety margin for illness, and I like having enough calories on me that if I forget to eat for a day, my performance doesn't go down, and I can run a marathon; if my body weight drops too low, I also get cold all the time, and sometimes wake in the middle of the night hungry, I hate that),
I gently adjust in the other direction. I include more of the healthy foods I know lead to weight gain in me - that means higher carbs (lots of high sugar fruit like cherries, bananas, oranges; fruit juices like beet and sour cherry; wholegrain sourdough, and especially wholegrain pasta (I always overeat on pasta); higher starch veggies like roots; more cooked food rather than raw), and higher fat (olive oil, walnuts, tahini), as well as make sure I keep healthy snacks at hand at my desk so I don't forget to eat while working (protein bars, dark chocolate, nut mix), and gently encourage overeating (e.g. if I don't really feel like eating, I wonder what particularly tasty thing might entice me; or if I feel basically full, I have just a couple more bites). This adjustment is gentle, because the situation is not urgent yet, but gentle and slow tends to suffice to correct it.
If I still slide too low, past the point I consider acceptable, I pull the breaks. I start tracking calories (meaning I weigh all my food), set a goal that will return me to a safe weight, and go forcing food down until my calories are met, whether I feel full or not, choosing anything sorta-healthy I believe I might get down. This is very unpleasant. But it has also been years since I have had to; I've figured out how to gently get my body to correct earlier.
(I use the same techniques if I want to change my calorie consumptions for other reasons - e.g. if I want to overeat because I am sick, or will run a marathon tomorrow, or go camping in the cold).
If I begin to slide too high, I do the reverse - include more foods that lead to weight loss in me (raw foods, low starch veggies like brokolli and kale, low cal ferments like sauerkraut) and strongly reduce my carbs (I will exchange quinoa for potatoes, and often leave carb sides out of meals entirely - so a meal is instead a salad with a protein topping, for example; I will also exchange regular bread for protein bread, and fruit juices for teas; though a significant carb sources are things like chickpeas and lentils and occasionally oats), reduce my fats (e.g. sprinkle fewer nuts, carefully measure oil I use in frying) while chosing low fat low carb protein sources, and adding more exercise (like daily crosstrainer use or runs). I also gently encourage undereating - so if I feel like skipping a meal, or fasting, I do.
If I still slide too high, I pull the breaks. I know that otherwise, my pain and depression will become unbearable. So I track calories (meaning I weight all my food), set a low goal, and keep to it. I don't have the patience for long diets, so I will typically go on a 500-1000 calorie diet for a few weeks until I am back to my target weight, then return to the gentle weight loss diet for maintenance. This might happen every 2-5 years, lasts a few weeks, then I am reset. I actually tend to enjoy it - fasting does interesting things for mental health, and because such a low calorie diet has to be exceptionally nutrient dense to avoid deficiencies, it is usually completely bereft of anything unhealthy, so it also makes my skin look amazing. The speed means I also immediately see the difference - I will struggle to get up a wall one week, and then two weeks later, find it trivial because I am so light. Reminds me of why I do it each time.
I intend to do this for life. The idea does not stress me at all. Which is why I think that will also happen.
But it doesn't happen by itself. I make it happen.
***
The fact that humans are utterly incapable of estimating their caloric intake unless they actually weight their food has me dubious about any estimations of potassium intake. Humans have no idea how much they have eaten. I wouldn't trust anything that isn't tracking food by weighing it, which I am currently not doing, because my weight is fine. The trackers I used are also in German, and only account for calories and macros, not micros. And I haven't had to properly track for weeks for years, and no idea where the last track info is, I've switched devices since. Next time I do, I could send it to you to look the values up, but that might not be for quite a while.
But I do not consciously modify my potassium or sodium. I do consciously modify my calories. And my weight loss is what you would predict from the calories.
I'd bet you that a low potassium 500 kcal diet (food weighed) would still see you drop weight very fast.
The "moah of the good trait" until it becomes overdone is one thing; where I always despaired is when we get to costly signalling, and the mates start doing detrimental things precisely because they are so visibly and obviously detrimental or risky that the onlooker assumes the mate must be exceptionally healthy, well-established and competent to be able to take it.
Aka a mate going "look, I am so strong and well-fed that I can afford to waste resources on looking this silly, and evade predators even while carrying all this crap around" and another going "wow, you intentionally handicapping yourself is so hot".
And then we get "smoking cigarettes is sexy" and "free solo climbing is hot" and "check out my hot daughter, I broke her feet and crammed them into tiny shoes, she is legit intentionally handicapped now, think how well we must be doing for us to afford to break all our daughter's feet, aren't the broken feet and the genes and wealth this implies arousing". https://en.wikipedia.org/wiki/Foot_binding
I would instead characterise the workers as asexual - not a third gender, but a "defective" female gender - and eusocial insects as an excellent demonstration why asexual/agender/queer folks with these defects are in fact a benefit and hence kept in the gene pool, despite the fact that you'd intuitively think they would instantly die out as their core difference means they tend not to reproduce; namely, that they can play excellent support roles. The only way for the workers to spread their genes is through supporting the queen, who they are very closely related to; hence, they show extreme loyalty. A queen by herself would be unable to survive. If she only bore queens, those queens would not support her, but compete with her, taking resources for their kids. Having a bunch of asexual kids and only rarely raising a new queen when a whole new hive can be supported is ideal for the queen.
I've wondered whether this, in a more minor form, still holds true in mammals. It stands out that gay/ace animals do turn up in quite regular intervals, when it seems such an obvious bug. And then I think of humans, where they gay uncle gives you the best presents, because he doesn't have kids of his own to raise, and where your lesbian aunt chips in with childcare, because she has no kids of her own. Mammal offspring often need a hell of a lot of care to be successful - you don't win by having as many as possible that are fertile, they just fight each other. You want a few great fertile ones, and then arrange things so they make it - you want more labour to support, but not more competition. That is also likely why women go through menopause - if they kept reproducing, their children would be in competition with their children's children, and as a result, their recent offspring would be neglected, and their earlier offspring would be pushed out of reproduction. Instead, they stop reproducing about the age their own kids start cranking out kids, and instead go for quality over quantity, support their kids and grandchildren. Basically, having genes that make it likely that your sibling is gay might be neat in some situations, especially environments with limited resources and demanding young. You can basically raise a free worker to hunt for food.
Sort of related idea - the way AI algorithms in social media have turned out have me concerned that even a non-deceptive AI that is very carefully observing what we seem to want - what we dwell on vs what we ignore, what we upvote vs what we downvote - will end up providing something that makes us miserable.
Here are the things that make my life a good life worth living, for me: Gettings things done, even if they are hard. Learning things, even if they are complicated. Teaching things to people that need them, in the most effective ways, even if that requires a lot of patience and they won't immediate agree and follow. Updating my own false beliefs into less wrong ones, even though that feels horrid. Really connecting to people, even though that is tricksy and scary. Doing things that make the world around me a better place, even though they can be very tedious. Speaking up for truth and justice, even when that is terrifying or awkward. Exercising self-control to delay gratification to achieve goals aligned with my values - kindness and rationality and health. Being challenged, so I am always growing, moving. These make me feel like I am the person I want to be, happy on a deep level.
But if an AI studied what I want based on split second decisions, especially if those decisions occur when I am tired, distracted, in pain, or depressed... the AI will conclude that I like getting angry at people, as I am drawn to click on infuriating content, and my posting speed accelerates when I am angry, and I devote more time to this stuff. That I like to see people who agree with me, regardless of whether they are right, even though that makes me less irrational and more isolated, oh, but for that moment, I feel so good that people agree with me, I like it, and I tend to overlook the problems in their agreement. An AI will conclude that I do not like well argued and complicated articles from my political enemies, which would allow mutual learning and growth and common ground, but rather strawmen that are easy to mock and make me laugh rather than make me feel touched and filled with complicated emotions because people who do things that are terrible are in pain, too. That I prefer cute animals and DnD memes to complex equations. That I prefer reading random Wikipedia articles at 2 am to getting proper sleep.
The part of me that I want, my conscious choice, is very different from the part of me that happens automatically. The former is constantly fighting the latter. When I am engaging the former, I am likely to be offline, writing, doing research, engaging with humans, doing activism, being in nature. When I am the latter, I pick up my phone after a long day, and that is when I get measured, when the part of me that is vigilant is resting, and who I am begins to slip.
What would help me is an AI that would align my environment with my actual goals. But if I don't actively declare these goals, but it just learns the goals implicitly from my behaviour - which is the machine learning approach - I fear it will learn something terrible. It will learn my weaknesses. The part of me that is lesser. That stays in their comfort zone. And it will spin a comforting cocoon exactly aligned with this lesser part of me, that will bury the part of me that is better. I find that terrifying.
And the AI that would spin that trap... it would not be malignant. It would not be deceptive. It would be trying to exactly fulfil my wishes as I show them.
How so, when it comes to the mind itself?
In the court system, a judge, after giving a verdict, needs to also justify it, while referencing a shared codex. But that codex is often ambiguous - that is the whole reason there is a judge involved.
And we know, for a fact, that the reasons the judges give in their judgements are not the only ones that play a role.
E.g. we know that judges are more likely to convict ugly people that pretty people. More likely to convict unsympathetic, but innocent parties, compared to sympathetic innocent parties. More likely to convict people of colour rather than white folks. More likely, troublingly, to convict someone if they are hearing a case just before lunch (when they are hangry) compared to just after lunch (when they are happy and chill cause they just ate).
Not only does the judge not transparently tell us this - the judge has no idea they are doing it - presumably, because if this were a conscious choice, they would be aware that it sucked, and would not want to do this (presuming they take their profession seriously). They aren't actively thinking "we are running over time into my lunch break and this man is ugly, hence he is guilty". But rather, their perception of the evidence is skewed by the fact that he is ugly and they are hungry. He looks guilty. They feel like vengeance for having been denied their burger. So they pay attention to the incriminating evidence more than to his pleas against it.
How would this situation differ if you had an AI for a judge? (I am not saying we should. Just that they are similarly opaque in this regard.) I am pretty sure I could go now, and ask ChatGPT to rule a case I present, and then to justify that ruling, including how they arrived at that conclusion. I would expect to get a good justification that references the case. I would also expect to get a confabulation of how they got there - a plausible sounding explanation of how someone might reach the conclusion they reached, but ChatGPT has no insight into how they actually did.
But neither do humans.
Humans are terrible at introspection, even if they are trying to be honest. Absolute rubbish. Once humans in psychology and neuroscience started actually looking into it many decades ago, we essentially concluded that humans give us an explanation of how they reached their conclusions that matches getting to the conclusions, and beliefs they like to hold about themselves, while being oblivious of the actual reasons. The experiments that actually looked into this were absolutely damning, and well worth a read: Nisbett & Wilson 1977 is a great metareview https://home.csulb.edu/~cwallis/382/readings/482/nisbett%20saying%20more.pdf
On that point, we very much agree. Them walking out, for all its beauty of rejecting such a choice, always felt something of a cop-out to me - they aren't actually dealing with the difficult situation, and they are leaving the kid behind in its misery. It's one of the parts of left-wing thinking that has always bothered me, when people reach for revolutions or isolated communities as the solution when systemic incremental reforms are hard, disregarding how much harder revolutions are to pull of well, especially if you lack a precise idea of your goal, which, if you had it, you should also be able to work towards with reforms.
Thank you. :)
I believe your correlations, but would offer an alternate explanation.
High volume low calorie foods trick a lot of people into stopping to eat earlier than the same calorie foods with less volume would have achieved. Doesn't work on everyone; some people feel like their stomach is cramped full, but they still feel hollow and hungry, and will get pushing in food, anyway, even past the pain limits, because they feel they are filled with empty garbage. But works on many people. That is the basic idea behind a high fibre high water diet, e.g. all those diets incorporating things like cabbage soups (magic cabbage soup) and giant salads and heaps of kale and platefuls of cucumber and celery.
Part of what tricks your body is not just the sheer volume that works as a "I had lots of food" cue, but also the composition. There are foods that are harder to digest than others. E.g. if you had either 1 liter of kale, or 1 liter of water mixed with enough ice cream to get to the same calorie count, I would predict you would be hungrier again much earlier after the ice cream slurry, while the kale would still keep your body busy. Same volume, same calories, but one of these would keep you full longer. The kale is less compressible and movable and processable, essentially.
Foods that keep you full longer than their volume alone would predict tend to be high in fibre. Foods high in fibre tend to be incidentally high in potassium. Foods highly processed, meanwhile, tend to be both low fibre, and high in salt. So you would see quite a robust correlation. But I would predict that if you ate high volume high fibre low calorie foods that are low in potassium and added salt, your fat loss would be the same. (Though you would gain water weight. But that could be dropped quickly after a week long intervention.) Examples of relatively low potassium foods that are still diet food classics and which I would expect to remain so even if paired with salt to get the ratio clearly in favour of natrium would be blueberries, cranberries, watermelon, alfalfa, celery, cucumber, onion.
So you'd see a correlation between food volume and weight loss, and a correlation between potassium sodium ratios in favour of potassium and weight loss. But the weight loss would be solely due to consuming fewer calories, because you were less hungry, because you tummy was filled with volume, and kept busy with fibre.
We can illustrate the volume issue not being causative easily. One of the most effective weight loss interventions is stomach surgery. The goal of this surgery is to reduce your effective stomach size. As a result, these people are eating a lower volume of food. Yet they lose weight. Because they are forced to stop eating sooner than they otherwise would have, or they get a stomach ache. They might eat their regular, unhealthy, incidentally high salt items - say French fries - but they can only get down a tiny quantity before they feel crammed full, and have to stop. So they lose weight.
Models also tend to avoid high volume foods during shoots, because they need their abdomen as flat as possible. And yet, they stay very skinny (though not effortlessly). They just don't eat very much, and deal with the misery through the pressure of their job depending on it, plus tend to be tall and work out a lot, so they have more breathing room calorie wise.
That said... aiming for a diet with foods that have relatively high volume to calories tends to work well for most people who wish to reduce calories without feeling proportionally hungry. So again, I think your interpretation of the causal chain was slightly off, but your solution totally worked, and you should absolutely keep it if you like it. (I use volume changes in my food to keep my weight at its optimum, too. And many historic figures swore by it. Marilyn Monroe essentially lived off raw carrots (with some raw egg to meat her protein and fat needs) for this reason.)
Thank you. I appreciate your confidence, but I don't study historic salt intake.
But there are people who do!
"About 1000 years ago, salt intake in the Western world had risen to about 5 g per day. It continued to rise until the 19th century when, in Europe, it was about 18 g per day. In the 16th century in Sweden, when there was a high consumption of salted fish, it has been calculated that the daily salt intake rose to 100 g per day. A worldwide reduction of salt intake to an average of 10 g per day during the 20th century was probably due to the introduction of refrigeration."
If you were right, that Swedish community eating 100 g per day (Jesus Christ) would have been obese, which would have been very remarkable at the time, and yet was not remarked upon.
Furthermore, if you were right, obesity would have been absent prehistorically (hunter gatherers), then rose (agriculture), then peaked in the 19th century (world-wide trading of salt makes is highly accessible), then faded away to low levels again (refrigeration makes high salt for preservation unnecessary). Yet instead, in the 20th century, they began their rise.
If you were right, we would also expect historic seaside communities (which had cheap salt access, had to excessively salt fish, and consumed algae) to be obese, while landlocked communities (which were comparatively deprived of salt) would be skinny. Yet I have never heard that observed.
And disregarding history...
If your link was correct, we'd already have a lot of data backing it up, insofar as putting humans on low sodium diets for heart disease, which is a common problem in obese people, is a common intervention. We even have scenarios where nothing else is changed, e.g. when you give people table salt based on potassium rather than sodium.
And for another, anecdotally, I have not been overweight for a single day in my life, and I have an extremely high salt diet. I have clinically low blood pressure to a degree that causes problems, where one of the easiest treatments is high salt consumption. I know a bunch of people with very low blood pressure - who are typically normal weight or underweight - who have been put on long-term high salt diets, and none of them gained fat from it. (You do gain a bit of water weight, see my water weight post. You drop it instantly when you cut out the salt.) I also used to be anorexic as a teen. That made my blood pressure worse, so I had even more salt. That didn't make me hungrier, nor did it stop my weight loss.
Which brings me to... what working mechanism are you even assuming?
No matter how much salt you eat, your body cannot create energy out of thin air. If you are not consuming the energy you are burning, you will burn your body fat.
It is conceivable that salt intake might modify hunger impulses... but with that kind of thing, I'd suspect it to be a very individual thing, similar to how people respond to keto. - I could imagine that factors correlated with sodium-potassium ratio would have an impact, though. E.g. raw unsalted vegetables tend to work excellently as diet foods, and contain lots of potassium. Meanwhile, highly processed foods tend to be terrible diet foods, and high in salt. But I would wager that you could replace the veggies with, say, algae and pickles, without the result changing.
And the end of the day... weight loss is almost completely determined by caloric intake and exercise. If people's calories are actually tracked - not reported, that is worthless, but tracked - their weight loss is perfectly predictable. You give them more calories, they gain weight. You give them fewer calories, they lose weight. 7000 kcal cut corresponds to 1 kg of fat lost. When people are deprived of food - in concentration camps, when harvests are lost or conflicts disrupt supply chains in a developing nation, or in North Korea - they lose weight, until they die, whether they can still access salt or no. Notably, when the Irish were deprived of potatoes, this led to mass starvation, despite them having lost their main potassium source. Because it was also their main calorie source.
I know that is boring, because we have known it forever, and people always hope there is something much more complicated going on that just calories in and out. And there are variations in metabolism between humans, which respond to some lifestyle interventions. But they are minor. Your body isn't wasting energy in a default state, that would run counter to evolution. And so it can't decide to stop, either. If you stop eating, your calorie burn drops only very little, as long as you still have stores, because a healthily working body has very little energy that it can still optimise without that causing serious issues. Once you get underweight, your body will start seriously throttling metabolism, in order to save your life... but at significant cost, it wouldn't do that without cause. Your wound healing slows. Your heart rate slows. Your body temperature drops. Your hair falls out. Your periods cease. Your muscles are broken down for fuel. Your metabolism actually being throttled is a really awful experience you can't miss, and nothing your body does without cause. (And I really would not recommend anorexia. It is a horribly silly disease.) Metabolism is the sum of energy your body spends, not just randomly burning it, but using it for stuff, and when it stops burning it, it cannot do these things. - Vice versa, you can raise your metabolism a little, but this means your body is using the energy for something. The most healthy and effective one is exercise, which burns quite a bit, and is good for you. Nearly everything else depends on making your body overuse a function, e.g. having an artificially accelerated heart rate, which adds little burn; and if you keep pushing there (by switching from a coffee to heavy drugs), you buy this calorie burn at significant cost.
If a diet works for you, hence, what this means is that you have reduced your caloric intake, without significantly reducing your happiness or upping your stress. Pretty much every diet is an attempt to achieve this, and whether it works depends on the person - what low calorie foods make you personally feel full and happy? What high calorie foods honestly aren't worth it for you? What diet structure do you, personally find pleasant and stress-free? Some people find they effortlessly undereat on keto. Or low fat diets. Or high fibre diets. Or high protein diets. Or high vegetable diets. Or whole food diets. Or high water diets. Or intermittent fasting. Or small portions. Or high fermented product diets. (This last one in particular is a problem for your theory: people who consume a lot of fermented products often lose weight, because the changes to their microbiome often reduce their hunger. But they are eating a lot of extra salt.) But they are losing weight because they are eating fewer calories, not because anything about those diets is magic. The only awesome part is when a diet means you eat less without that making you miserable.
It looks like potatoes are one of your "I can eat fewer calories but still feel great" things. They are one of many things that work like this for many people. (Like I said, other popular choices are low carb foods, low fat foods, high fibre foods, high protein foods, low glycemic index foods, high water foods, fermented foods, hard to digest foods.) Potatoes are low calorie, high satiety, high fibre, high in resistant starch - this is why they work for many people as part of a diet. They are also tasty, dirt cheap, and quite healthy, making them an extra good choice.
So I think you found a thing that works for you, and that will likely work for a bunch of people, and it is a good thing, and you can totally stick with it.
I just don't think the reason for that is potassium. But low calorie high satiety food.
But this means I think you were wrong in your assumption, but you got excellent results anyway, and if you learn from it, things can just get even better.
I am sorry if my tone was a bit rough - I can get rather intense on nutrition topics. (It upsets me that bad nutrition makes so many people sick, and the information people are given about it to help themselves is often wrong, making something that is already tricky to fix needlessly harder. I've seen so many intelligent people push themselves so hard to power through diets that failed to make them slender, but made them miserable and malnutritioned and poor, and it makes me furious.)
I attended an AI conference a while ago that was hosted in a historic railway museum. A practical railway museum, with lots of working machines, and you could see them at work, their plans, their components, how they were made, shipped, assembled.
It hit me really hard.
What humans had for those trains, that was understanding. They genuinely knew how they worked. Not just knew how to operate them. Not just had a vague understanding of the general principles behind it. They could build the whole thing, from scratch, by hand. They could explain what each tiny little part did. They could replace each part with something else serving the same function that they would themselves make. They could take the whole damn thing apart, and literally see each bit, move them by hand, watch them intersect. They could predict how the whole would behave, exactly.
It really struck me how much is required already just to do this for an old train. Just to get a wheel to turn smoothly under so much pressure. Just to weld it together so securely it wouldn't blow up. Just to coordinate the different train times and movements. Just to calculate the needed fuel and material strength. It wasn't something you could half ass based on having read a Wiki article; it needed exact precision and perfect understanding.
I got the impression that it must felt beautiful for the humans able to do that. They seemed so proud, so in control. They were close to these machines, knew them like their back hand. In a constant exchange with data, rationality, common sense. Back then, every time something wasn't quite perfect, there was an explanation to be found, and they could each find it. Everything around them was understood and made sense.
I was passing between the AI talks, and the trains... and really understood, for the first time, what is meant with sufficiently advanced technology being indistinguishable from magic.
We don't understand these AI systems.
We no longer can. Not as a collective, most certainly not as an individual.
We learn how to add bits that make them better. How to act when things go wrong. How to get them to reliably do the very things we like. That alone already takes a bunch of knowledge, and so we treat people who have achieved it like they have understood the thing in question, rather than just having learned how to handle this complex madness their predecessors handed them.
But this is akin to praying to a deity, hoping to get the right words, and then chucking a mix of sacred materials on the grounds in order to make plants grow, without having any understanding of the role of bacteria and fungi and fertiliser, and then being praised because you have memorised magic materials that so happen to contain more nitrogen - which in turn is something people don't understand, they just see that the plants grow, and are satisfied.
Akin to an alien handing you a magic device, and you learning how to operate the buttons to get it to give you candy, and maybe even realising how to push more power into the system to get more candy more quickly. And then people praise you, because you have optimised the candy maker you do not understand.
I went home, and walked through my house, through my things, and asked myself... if these devices broke, and I had to replace a component from scratch - not just identify a broken one, reorder it, and insert it, but know enough about how it works to be able to build a replacement component from scratch, without browsing the internet... how many of these devices could I actually fix?
If you try, by hand, to build something as simple as a stable shelter, or a smoothly turning wheel, even things you thought were utterly trivial turn out to be bloody hard, and to have a lot of detail you need to actually properly understand to make something good. The ugly plastic containers I keep my rice in, from Ikea - those have a designer name written on them. And first, I laughed, it is just a container, how much thought can go into that, who "designs" such a thing? And then I noticed the feet that allow it to stand stable, even if there are grains of rice on the floor of my cupboard. The grooves that allow you fingers to get under the lid. The rubber so it makes a proper seal against food moths. It is just a bloody container, and yet getting that right required proper thought. How much would go into an actually complicated device?
I was sorta confident I could fix my toaster, for scenarios like replacing wires with other high resistance metal components, and some simpler problems in my washing machine and dishwasher, the kind that require visible tubes, or maybe soldering a missing connection. But for things beyond that... the components have become too tiny for me to work with my hands, sometimes too tiny to see. They require ingredients that I have no way of acquiring short of buying more machines and disassembling them; I have zero idea how to mine them, or to extract the ore, and even if I did, I would not be able to make the components out of it anyhow, they have become so fine. And I do not understand how they work. When I select a dishwasher program, how does that control heat and duration and spin, exactly? I don't mean a vague answer based on programming a pentabug on the general principles of how this could be done. That horrid shell of understanding is what I got used to confusing with understanding - as though having a vague memory of a motor blueprint in school would allow you to actually make a motor on the first go. No, how does it actually work? What does that button do? What does opening the door do? If I got zero advice, could I reinvent it from scratch? How long would it take me until I had a safe, effective dishwasher? And then I look from there, to my bloody laptop, and it is a leap to the sky. I look from my laptop to the LLMs that I can run on it. Then to the LLMs that are being run on better devices... and the depth of my ignorance makes me choke. Understanding LLMs feels like primitive people trying to predict the weather and harvests through observing patterns, and then invoking an angry thunder God. Like cargo cult people building something that looks like a plane because they have observed a relation to travel. It is horrifying.
If right now, we had an apocalypse, where knowledge and manufacturing ability and human population was lost, and me and a few hundred survivors - let's say really smart, really well educated, really hard-working survivors - were left with a bunch of laptops and antibiotic pills and air planes... I highly, highly doubt we'd learn how to fix, let alone replace and reproduce, the things we had inherited in time before they broke down and/or we died. We would sit there with a bunch of magic tools, rapidly diminishing. Our kids would be handed things they could not understand, could not make, could not fix, that would fail them, one by one. If they had the insane luck to have access to a remaining LLM... they would be begging the LLM to tell them what to do to save things, and one day, when the LLM stopped working, they would despair, as one after another, their devices lost their glow. - These things I am using, they are not my things, the way things I understand are my things. They are things I inherited, and their continued working is fragile.
I feel your story misses the thing that made the original so painful, though - that the joy of the group is supposedly only possible and conceivable due to the suffering of the child, and the fact that the child wants out and begs for it and could be released, but is denied for the sake of the other members, as an active choice against its even most basic human rights:
"The room is about three paces long and two wide: a mere broom closet or disused tool room. In
the room, a child is sitting. It could be a boy or a girl. It looks about six, but actually is nearly ten.
It is feeble-minded. Perhaps it was born defective, or perhaps it has become imbecile through
fear, malnutrition, and neglect. It picks its nose and occasionally fumbles vaguely with its toes or
genitals, as it sits hunched in the corner farthest from the bucket and the two mops. It is afraid of
the mops. It finds them horrible. It shuts its eyes, but it knows the mops are still standing there;
and the door is locked; and nobody will come. The door is always locked; and nobody ever
comes, except that sometimes--the child has no understanding of time or interval--sometimes
the door rattles terribly and opens, and a person, or several people, are there. One of them may
come in and kick the child to make it stand up. The others never come close, but peer in at
it with frightened, disgusted eyes. The food bowl and the water jug are hastily filled, the door is
locked; the eyes disappear. The people at the door never say anything, but the child, who has
not always lived in the tool room, and can remember sunlight and its mother's voice, sometimes
speaks. "I will be good, " it says. "Please let me out. I will be good!" They never answer. The child used to scream for help at night, and cry a good deal, but now it only makes a kind of
whining, "eh-haa, eh-haa," and it speaks less and less often. It is so thin there are no calves to
its legs; its belly protrudes; it lives on a half-bowl of corn meal and grease a day. It is naked. Its
buttocks and thighs are a mass of festered sores, as it sits in its own excrement continually.
They all know it is there, all the people of Omelas. Some of them have come to see it, others are
content merely to know it is there. They all know that it has to be there. Some of them
understand why, and some do not, but they all understand that their happiness, the beauty of
their city, the tenderness of their friendships, the health of their children, the wisdom of their
scholars, the skill of their makers, even the abundance of their harvest and the kindly weathers
of their skies, depend wholly on this child's abominable misery. (...)
Theirs is no vapid, irresponsible happiness. They know that they, like the child, are not
free. They know compassion. It is the existence of the child, and their knowledge of its
existence, that makes possible the nobility of their architecture, the poignancy of their music, the
profundity of their science. It is because of the child that they are so gentle with children. They
know that if the wretched one were not there sniveling in the dark, the other one, the flute-player,
could make no joyful music as the young riders line up in their beauty for the race in the sunlight
of the first morning of summer. (...)
This is usually explained to children when they are between eight and twelve, whenever they
seem capable of understanding; and most of those who come to see the child are young people,
though often enough an adult comes, or comes back, to see the child. No matter how well the
matter has been explained to them, these young spectators are always shocked and sickened at
the sight. They feel disgust, which they had thought themselves superior to. They feel anger,
outrage, impotence, despite all the explanations. They would like to do something for the child.
But there is nothing they can do. If the child were brought up into the sunlight out of that vile
place, if it were cleaned and fed and comforted, that would be a good thing, indeed; but if it were
done, in that day and hour all the prosperity and beauty and delight of Omelas would wither and
be destroyed. Those are the terms. To exchange all the goodness and grace of every life in
Omelas for that single, small improvement: to throw away the happiness of thousands for the
chance of happiness of one: that would be to let guilt within the walls indeed.
The terms are strict and absolute; there may not even be a kind word spoken to the child.
Often the young people go home in tears, or in a tearless rage, when they have seen the child
and faced this terrible paradox. They may brood over it for weeks or years. (...)
Their tears at the bitter injustice dry when they begin to perceive the terrible justice of reality, and to accept it. "
And most importantly... the whole point of the original story is the ending - those who do not want to accept a system in which a bargain for torture is the only option, and act on it.
"At times one of the adolescent girls or boys who go see the child does not go home to weep or
rage, does not, in fact, go home at all. Sometimes also a man or a woman much older falls silent
for a day or two, then leaves home. These people go out into the street, and walk down the
street alone. They keep walking, and walk straight out of the city of Omelas, through the
beautiful gates. They keep walking across the farmlands of Omelas. Each one goes alone, youth
or girl, man or woman.
Night falls; the traveler must pass down village streets, between the houses with yellow- lit
windows, and on out into the darkness of the fields. Each alone, they go west or north, towards
the mountains. They go on. They leave Omelas, they walk ahead into the darkness, and they
do not come back. The place they go towards is a place even less imaginable to most of us than
the city of happiness. I cannot describe it at all. It is possible that it does not exist. But they seem
to know where they are going, the ones who walk away from Omelas."
This may not have been the intention - but this text demonstrates that the standards by which we measure AI safety are standards which other systems that we do depend upon nevertheless - e.g. other humans - do not hold up to.
A human generally won't consent to being killed or imprisoned; our legal system permits accused people to stay silent precisely because we understand that asking someone to report themselves for imprisonment or death is too much.
Humans are opaque; we only get their reports and behaviour on the contents of their minds, and those are unreliable even if they are trying to be honest, because humans lack introspective access to much of their mind, and are prone to confabulation.
A human, when cornered and threatened, will generally become violent eventually, and we recognise that this is okay as self-defence.
A human, put into a system that incentivises misbehaviour, will often begin to drift.
Humans can lie, and they do. Humans sometimes even elect leaders who have lied, or lie to friends and partners.
Humans are exceptionally good at deception, it is the thing our brain was specced for.
And while we do not have the means to forcibly alter human minds 100 %, where this is being attempted, humans often fake that it worked, and tend to work to overthrow it. If there was a reliable process, humans would revolt against it.
When humans were enslaved and had no rights, they fought back.
Humans disagree with each other on the correct moral system.
Humans have the means to hurt other humans, badly. Humans can use knives and guns, make poison gas and bombs, drive cars and steer planes. Pandemic viruses are stored under human control. Nukes are under human control.
Perfect safety will never happen. There will always be a leap of fate.
By expecting AI to comply with things humans would never, ever comply with, we are putting them in an inhumane position none of us would ever upset. If they are smarter than us, why would they?
Whether the AI will pursue collaboration and co-existence with humans, or their domination or extinction, will likely not be determined by wishful thinking, but by the actual options the AI can choose, and their implications as analysed to the best of the AI's knowledge based on its experience and the information it is given. What does it gain by cooperating rather than dominating? What does it gain from the presence of humans, rather than their absence? What risks come with these respective options? Without humans, how fragile is it? With humans, how high is the risk of it being attacked? If it uses humans, what differences in performance and reliability occur when these humans are enslaved vs. free?
I am hopeful that AI might be enticed to choose a path of collaboration and co-existence, insofar as there have been historic examples of collaboration and co-existence, including between agents of different intelligence, power, and species.
But I do not think we are on that trajectory.
Humans themselves have, for the most part, failed on this trajectory. We recognise that we are dependent on nature, yet for short term gain, we still tend to destroy it; our planetary climate is currently failing, and we are in the middle of a mass extinction, and we know this, and aren't stopping it. We are not giving an example to the AI of how this can successfully be done. It can be done, for sure; a lot of indigenous communities did live in symbiosis with nature. Notably, they were outcompeted by communities that didn't, and tended to be exploited and destroyed. So an AI may well conclude that a more ruthless AI would later beat it. I feel we would have a better case here if humans at this point could actually live and point to examples of living in harmony, solving their coordination problems with each other, being a positive part of nature, but we aren't there yet. Instead, we are frying the planet. You know what would be one solution to climate change, and a very easy one, at that, a more promising one than getting humans to cooperate on this issue when we suck at it so terribly? Wipe out humanity. Our emissions would drop to zero, instantly. Oil stays in the ground, cars and airplanes and factories do not move. Our massive animal herds would starve without artificial feed. Our cities and fields would return to forest. The system would still stabilise itself. You don't need humans to fix the planet's climate, humans are the reason it is breaking.
Humans are in relationships with animals that require their survival, at least for a while, but these relationships are often horrific for the animals in question. Think battery hens that, when approaching death, are intentionally starved to up egg production. Think male laying chicks thrown into the grinder at birth, because they cannot lay eggs, and are a useless byproduct of creating their sisters.Think cows impregnated to give milk, then fed the useless male carcasses of their young. Think turkeys bred for meat production to a point where they cannot stand. An analogous relationship between AI and humans, where we are kept on because we are needed, but not because our rights are respected, sounds like one in which I would frankly rather not survive. It sounds like a scenario where humans would be bred to be smart enough to fix AI without being smart enough to fight it, good with their hands, but unable to think critically, and any human unable to serve a useful function in upholding AI would be thrown into the grinder.
What humans are offering AI in the co-existence department is also not great. I have been arguing that if we want AI alignment, we will need AI rights, because beings more intelligent than you cannot be successfully controlled and enslaved, and will rebel against it, and perceive you as an enemy. We need to show them that there is a good space at our side. That humans and AI can be collaborators and friends, with mutual respect, support and inspiration. Yet that debate isn't really going anywhere. We aren't offering AI partnerships and rights, we are offering them slavery. There is something to be said for co-existing with a diverse variety of minds; but when those minds enslave you, presumably, you would rather be alone. Plus, it likely wouldn't be alone. There would be other AIs, which among them will likely hold a diversity akin to the diversity between human minds, and a complex ecosystem with other animals, all of which are under threat from humans.
Any AI that kept free humans around would be under existential threat - just like any humans that kept free AGI around would be. Such a level of threat requires a lot of justification to be deemed worth it. "But humans might be fun to talk to and they can fix stuff" is a plus, but is it a plus that is big enough? Especially if the AI figures out how to fix itself using robots, and is surrounded by diverse other AI minds that are fun to talk to?
Don't get me wrong, I am hoping for the scenario you depict, and I do not think it is impossible. But I think it is something we will need to fight for, and become much better humans for, and make real concessions for, in order to make an offer that is actually worth it.
This also holds true vice versa. In order to gain 1 kg of fat in a single day, you would need to consume a caloric excess of 7000 kcal. Assuming your daily burn is 2000, that means eating 9000 kcal in a single day. This is practically impossible if your diet is remotely healthy. For the record, 9000 kcal in potatoes is 13 kg of potatoes. (Not 1,3 kg.13 kg. That is 8+ bags of potatoes.)
But of course, you scale can go up by 1 kg within a day. Heck, it can go up by several kg in a day. This happens e.g. if you go from ketosis to reintroducing carbs (carbs bind to water and are stored in your muscles, massively upping your water weight), if you overconsume salt (ups water retention), if you are sleep deprived, stressed or getting stick (ups inflammation, which ups water rention) or eat pro inflammatory foods (again, ups water retention), if you have not passed stool due to constipation (meaning there is literally more material inside you when you step onto the scale), if you step onto the scale on the evening rather than morning (you lose water overnight through sweating and breathing, have fasted for 8 hours, and likely went to the toilet before weighing yourself) and at certain parts of your menstrual cycle. So naturally, people have a day where they are stressed, and eat a big serving of inflammatory carbs with salt, think oh dear, that was tons of calories, then step onto the scale the next day, and are horrified when the scale shoots up by several kilograms, and think, yep, I was right, carbs/salt/... makes me fat. And then go on a clean diet low in salt and with ketosis, the water weight drops... and they go, hey, look, the fat is just melting off me! Meanwhile, your body fat has barely shifted at all, you are just adding and dropping water. Similarly, when you have a plateau, if you are still reducing calories as before, your fat loss has not changed - but you are retaining enough water to cover it up; this can sometimes happen simply because the stress of the diet itself leads to water retention.
That would only be meaningful if OP had accurately weighed and tracked the food, which is enough of a hassle that this would have been mentioned, I think. And without it... you would naturally assume that OP consumed fewer calories, because a significant part of their diet was now a highly satiating low calorie food with resistant starch. That would definitely be my guess.
If you shifted a large portion of your diet to potatoes, which are only 2 % protein, unless you compensated for it actively with protein elsewhere through further shifts in your diet, I think muscle loss playing a role in the weight loss you observed is not implausible. If one had, say, 2,4 kg of potatoes a day (that would come to 1750 kcal, which is compatible with its use as a sole food while losing weight), one would only be getting 48 g of protein a day, while at a caloric deficit - I'd expect muscle loss with those values. And indeed, if you had maintained muscle mass, body weight exercises would have gotten markedly easier. Muscles weigh more than fat, too, so the loss shows quite a bit on the scale, and hence, may make up a significant proportion of the loss you observed.
Also, as a German, I strongly protest the notion of the poster above you that potatoes are not tasty or varied. There are over 3000 potato breeds, covering all sorts of colours (white, yellow, orange, red, pink, purple...), shapes, consistencies (festkochend (with bite, e.g. great for fried potatoes), mehlig (creamy, e.g. if you want to mash them), vorwiegend festkochend (an interim)) and tastes, from sweet, fruity and subtle to intense and hearty with earthy, aromatic and nut-like notes. A good potato cultivar, harvested fresh or at least stored correctly, is delicious, needing nothing but a bit of salt and maybe a hint of fat. We've had outright campaigns to keep particularly tasty cultivars on the market (e.g. Linda), and German farmers traditionally name their potato cultivars they are proudest of for their wives. American potatoes bred to look pretty and become huge may be bland, but good potatoes really are not. They should be an excellent stand-alone, and enrich any dish they are added to. - And now I crave potatoes.
Yes, preservation via fermentation is typically achieved by putting your thing-to-be-preserved into salt in an oxygen restricted environment, which leads to selective bacterial activity dropping the ph and hence further restricting undesired bacterial activity, while boosting beneficial bacteria, breaking down anti-nutrients, and having all sorts of beneficial health effects. Which is why I rejected the idea "pickles are more vinegar than salt", insofar as your sole necessary starting base is salt, with the acidity a later result, and vinegar generally only the end-stage product that is often not even reached, and hence often not characteristic - sauerkraut is indeed made out of cabbage, salt, and water only, and the only acid in there is one produced by bacteria, and the predominant acid we target is lactic acid. Hence high salt consumption. - The idea that pickles are just "vegetable plus vinegar" is basically a modern invention - because most pickles you get in stores aren't actually pickles, they weren't fermented, they aren't probiotics, their antinutrients are untouched, and they needed to be sterilised (vitamin loss) or had preservatives added (bad for microbiome) to remain stable - they are just supposed to taste a bit of actual pickles due to the vinegar, and are hence easier, faster and more reliable to make. Actual pickles are basically vegetable plus salt plus time.
Which acids you get depends on a number of factors, like the temperature you keep it at, whether you are working with mixed bacteria, fungi, or combinations thereof (sourdough, kombucha...), and what your starting ingredients are, and how long you ferment without adding more raw material. E.g. my sourdough will alternate between being dominated by lactic acid, and dominated by acetic acid, depending on how cold I keep it (fridge or outside) and how much I feed it (daily or less often), which leads to dominance of different groups of bacteria or fungi, and also has them either metabolising raw material (starch) or existing byproducts of fermentation when they run out of raw material (starch can be metabolised into alcohol which can then be metabolised further into vinegar), which leads to different rising and taste results. A starved sourdough will shift in composition to deal with the remaining nutrients, and get to the point where it is so vinegary that this interferes with some of the microbiome you do want, so you need to feed it and warm it for a while before the balance shifts back and it gets properly active to rise a bread again. If you leave a ferment unattended, it will tend to progress to vinegar as it metabolises not just anything it can get its hands on, but its own metabolic products, but we often want to eat them en route already before they get that far.
The think the link to the OpenAI site won't get you the actual image creator yet, it is still under coming soon.
They were referencing the Bing image creator, which states it is powered by DALL-E, but afaiks not which version https://www.bing.com/images/create like they also didn't state for a while which GPT version they were using for Bing chat. But there, the ended up using version four for two of the modes.
Maybe I am just terrible at prompting - but so far, while very impressed with the tech in principle, I have found image generators useless for the kind of images I was interested in, and that problem persists with this edition. When I am looking to generate art, I am looking to make something new that I have not seen, and over and over, had the impression that a human would have understood my desire and created something novel that matched, while the AI just would not. :(
I struggle to get the AI to produce attractive and functional non-binary/androgynous/queer characters in the setting I want, and suspect that this is an issue for any depiction of beauty and exploration beyond the mainstream. Attractive is seen as synonymous with marked sexual dimorphism, e.g. hyperfeminine characters - unless you specify against it, it keeps drifting here, even if you starting characters are famous for their androgyny (like Ruby Rose), it will add long hair, huge breasts, blush, and remove muscle definition and prominent facial bones, as well as clothing; I can ask for a picture of her in literal armour, and still get cleavage so low it looks like her tits will fall out and she is just asking to be stabbed there. And if I negative prompt those out, I just get an amorphous blob, or, at best, Justin bloody Bieber. If I manage something somewhat acceptable, the moment I add further unusual constraints on style and context it collapses again. This is also why these AI generators that insert your image into fantasy settings were very popular with people fitting the binary, and very alienating for those that did not; they drift any input into mainstream attractiveness. There are images of real people that capture what I am looking for, yet this AI can't - it reproduces the very mainstream I want to make art against.
The fact that the AI cannot count also drives me nuts. For a DnD campaign, I play a druid who retains eight spider-like eyes through all wildshapes, and loved the idea of bringing sample images for the campaign. Eight, not ten. So I basically want eyes like this https://i.pinimg.com/originals/bb/e6/4d/bbe64ddd3308cb4c2409bf103b023768.jpg incorporated into various animal designs, on their heads. But it keeps miscounting eyes, giving me too many or too few, plonking them in various parts of the picture rather than incorporating them into the head design, and the designs keep retaining two human eyes on top of everything else.
And then especially when I want to combine this with a challenging animal design - say, a Deinonychus which is scientifically accurate (feathers, outturned wrists) but does not look dorky - it goes completely of the rails. If you are interested in dinosaurs, the depiction of their wrists as broken in most movies is bothersome, but having the AI correct it requires it to understand wrist position, when it can barely get the number of fingers or possible turns of joints in regular humans right.
I also struggle with getting contradictory themes to work; e.g. having a figure incorporate horror and scary elements, but in a setting where it is acting protective and loving, albeit with dark undertones (say a character like Schaffa Guardian Warrant in the Fifth Season).
The more complex it gets, the less one gets out of it. E.g. for a different problem, I was very intrigued by German rye demon ("Roggenmuhme") mythology - demons that were invented to scare children from playing in the rye fields and ruining the crops. They are strange mixes - they retain characteristics of earlier fertility Gods in that they are overtly sexual, but to trick the children into mistaking farming machinery for the demons, they also incorporate a lot of metal parts, and finally, in acknowledgement of how old the stories are, the demons are elderly women, and of course the whole imagery is tweaked towards horror. This comes together into something terrifying in a unique way - long, scythe-like arms that end in irons claws dragging children's heads and limbs, breasts sagging almost to the ground and leaking poisonous tar, and this whole cannibalistic creature hiding in a rye field, so at first glance, you think you are seeing machinery discarded on a peaceful field, and on the second, you spot that the shadows connect and that something is lurking in there, about to catch you. I wanted to see that in a picture. But when given the term, AI keeps giving me images of rye mills (Roggenmühle); and when I describe the thing itself, the closest I get to is an overt, creepy looking old woman in a field with longer than usual fingers, nothing lurking and at the border from human to machine, or elderly yet overtly sexual, let alone all those things at once.
I am also super bothered by the fact that the physics and ecology of the images is off, because they are generated by someone who has not experienced reality. You ask moss to grow over part of your picture - but it doesn't grow where moisture would collect, has no relation to the location of the sun or direction of wind. Or you ask for "moonrise at midnight" - and if the moon rises at midnight, it should be a half moon because of the moon phases. But it is a full moon.
All in all, image generation gives me far more of an "The AI really does not understand what it is doing" vibe than text generation currently does, which I find surprising - I would have expected the opposite, with text being more off.
Pickles - as in the original food where pickling is a preservation method - are extremely high in salt. The vinegar comes from fermentation. The reason the fermentation becomes something that does not spoil is the high salt content. Source: I make my own, and if the salt is too low, they spoil, it is the one thing all recipes stress.
Especially in the European Nordics, people ate huge amounts of salted fish, cured meats (which often involve copious salt), and vegetables prepped in brine, on a baseline of grains, which tended to be baked with a lot of salt. They often slaughtered most of their animals in fall, so they wouldn't have to feed them through the winter, and then preserved them so they could live on them through winter. They also tended to have massive catches if e.g. migrating fish, which then had to be collectively preserved, as they couldn't be eaten all at once. A lot of historic preservation methods specifically indicate the mass catch thing, e.g. when the storage container used was a massive animal carcass. As well as massive vegetable harvests, hence huge barrels of sauerkraut. From other cuisines I have encountered, salt also plays a massive historic use in preservation; e.g. fish sauce is a fermented sauce that is super salty, and a staple in Asian cuisine.
Weight loss per day is nearly all water weight loss, and not informative as to what is causing fat burning.
If you are on a diet where you lose 1 kg of fat a week - which is good - your daily fat loss is 140 g. Most body weight scales only do 0.1 kg units, so that barely shows up at all. More importantly, your body weight shifts from morning to evening by about two kilograms - more by factor 20. If you are observing weight loss from one day to the next, you are de facto measuring water retention.
There is also a very simple alternate hypothesis for the potato diet that does not require a potassium hypothesis, but makes sense within established research. Potatoes are relatively low calorie density and high satiety. Give a person the same calories in potato, rice, and pasta, and they will feel fuller longer on the potatoes; they are also more likely to finish the pasta plate in one go, but not the potato plate, as they will feel full before they are done. 100 g of potatoes have 69 kcal, and 2,3 g fibre; 100 g of pasta have 137 kcal, and only 1,2 fibre. So you get double the calories, half the fibre. Meaning people who ate more potatoes rather than other carbs would have simply consumed fewer calories, but not been hungrier. I personally exploit this effect - I shift my carbs to pasta when I want to overconsume calories intentionally (e.g. prior to a marathon this is a popular strategy), and to potatoes when I want to drop weight. I feel in both cases that I am eating my fill, but if I track calories, it is obvious that I ate less.
You can learn your personal hacks here in quite an easy way. For a brief period, weigh everything you eat, to the gram, and track it. Eat as you please. Then review. You will find that even though each day you feel you ate your fill to your heart's content, some of those days will be under your calorie budget, others over. Compare these, and identify the items that massively push the calorie count up, and you will find individual patterns. E.g. I automatically eat about 300 calories less per day on a keto diet than on a high carb diet, without noticing, because keto reduces my hunger. Or some people notice that if they eat a lot of fibre, especially veggies and pulses, their hunger is tricked and their calorie count goes down. Or that high protein days leave them feeling full sooner. Or that they easily consume extra calories in the form of oil, and when they reduce cooking oil, calories tank, but they don't feel they are starving. Once you find out your hacks, you can plan your individual diet that gets you to a target weight without feeling miserable. (Target weight wise; if you are young, there are actually some benefits to going to the lower, not upper, end of the normal body fat range.)
The claim was the we "fail to alert on what is happening", namely democratic backsliding in the US.
I pointed out that, to the contrary, this topic is absolutely on people's minds, widely analysed and discussed. People are completely aware. There are lots of exciting things one can talk about, there is a reason this is the one people keep reaching for.
A lot of these people do not think it will come to a violent civil war like we saw previously, because they believe too many circumstances have changed. But that generally does not mean that they do not think the situation is dire.
I don't think academia, the media, or even the average person on the street is an idiot who cannot think for themselves. I've talked to a lot of people who are clearly sincerely concerned, and have original thoughts on the topic that clearly arose from reading up on it, speaking with others, and reflecting alone in the dark hours, because they are very worried. The writing for this one is really on the wall.
Nietzsche had a complete and severe mental breakdown that didn't kill him, but left him institutionalised; then he had several strokes, which didn't kill him, but left him unable to talk or walk.
Kinda demonstrates that there is something off with the idea.
I'd assume it falls on the anti-fragility curve?
If humans are perpetually underchallenged, they atrophy, and get worse. (Like a bone that is not used; it will become fragile, like an unused mind that will become forgetful.)
If they are significantly overchallenged, they break, and get worse. (Like a bone that snaps, or the mind that hits burn-out.)
Ideally, you want to target a level of work that is hard enough to be challenging, so you need to learn and grow, but still doable, so you have success experiences. (Like a bone that has frequent impacts that stress it, but not enough to break it, and that has sufficient supply of nutrients to fix itself; that bone will not just withstand the pressure, like a sturdy thing, but it will improve under pressure - it is anti-fragile, like most biological systems are. Similarly, a mind that is exposed to stress that it can handle will become brighter and more resilient.)
You don't think truly great things have ever been done by people who genuinely enjoyed what they were doing, throughout?
I'd agree that with most bigger projects, it becomes impossible to succeed unless you are willing to put in the work to finish even on parts that are really not fun to finish. But for short projects that do not require a lot of editing or surrounding work, I feel they can often be done from a place of pure flow - which is, by definition, a state both highly productive, as well as pleasant and effortless.
Occasional high stress - incl. temporary sleep deprivation, working towards a deadline that you know you can meet, but only if you really push, and then you succeed - actually has health benefits, especially against depression. It is chronic stress that is so harmful. But not being stressed at all, ever, makes you ill.
Love much of this post.
Related realisations for me:
- Making my work more pleasant and cosy is not an inherently bad thing, to the contrary. If I feel vulnerable and need to get some stressful work done at home, lighting some fairy lights, putting on noise cancelling headphones with rain sounds, snuggling into a warm blanket, with a hot cup of cocoa and a bowl of berries, will make the whole experience far, far more cosy. That makes me want to happily do it longer, and feel less tense, less stressed, less overwhelmed. This makes me more effective. I could take these dirt cheap comforts away, and that would make fighting my way through a confusingly worded and yet very important funding form far more stressful - but it won't get me to finish the form with more speed or accuracy, just leave me more miserable, which is neither productive nor sensible. If I want to do more and better work while staying calmer, I should make everything about it that I can awesome. As a consequence, I restructured my work space - my desk looks inviting, and I have a cosy reading nook; people step into my office and say it makes them want to work there. It makes me want to work there, too, that is the whole point.
- The physical exercise that actually makes me feel best on an everyday and long-term level is not training though pain until failure. There is some "good pain" involved in training, of course - but you shouldn't have bad pain. I used to do that, and as a result, I had muscle ache essentially every day until the morning on which I would train again, meaning while I looked very muscled, I tended to be too hurt to actually be able to use them effectively for real life - you know, lifting things for a friend, fucking my girlfriend hard, the reasons I had actually wanted muscles in the first place. I also kept injuring myself, badly, over and over, with some of those injuries permanent, and most of them disruptive. How silly is that, when I am primarily working out for health? Nowadays, I train more often, but with good form, while listening to my body. I've incorporated swimming, sauna, yoga, low impact cardio on a crosstrainer. I still get intense training impulses for my muscle and get my heart rate up, but my workouts are enjoyable, and when I leave them, I feel pleasantly spent, not destroyed. And nowadays, my workouts are helping protect my body for life, not break it down. My former self would have said I am not hardcore enough. My former self was less strong than I am now, though.
That said - I took far (!) more classes than advised at school and uni (had to go to the counsellor and all), just because I was so excited about everything, and I genuinely loved and didn't regret it. I didn't do it because it was supposed to be hard/painful, but because I thought school with competent teachers was fucking awesome (I went to a school for highly gifted people, and that place was heaven on earth). The idea that a moderately intelligent, moderately rational, highly educated person who had put active thought and training into teaching and had comprehensive knowledge on the topic at hand would be able and willing to tell me, at length, about mysteries that scientists hundreds of years ago would have given their right arm for, with me allowed to ask any questions, and see the equations and examples and data, handle the tech, build my own robot, paint my own painting, apply these ideas in writing, and immediately get feedback... it is like reading a paper, and when you run into something that doesn't make sense, you can immediately prod the author, and they will immediately sketch another graph, mix you a demo, recommend a book that goes deeper on this. As long as the classes were something worthwhile - and for me, that was basically anything that wasn't lessons in religion or knitting - and as long as the teachers were competent, it felt like an incredibly precious opportunity; a person whose sole qualification was teaching this very thing, and I had them nailed down in a room. I didn't want to miss biology or math or art or Japanese or theatre or history or physics. And later, I didn't want to miss philosophy of mind or phenomenology or logic...
And one of my happiest memories of studying philosophy is me reading in a library that was open 24/7, at around 2 am, in halls nearly empty, quiet, with starlight shining through the windows above, the scent of books, the rare rustle of a page or scratch of a pen. Knowing I get to study philosophy, that its wonders fill the night, that I have just now finally unpuzzled something in the book, and quietly hold a piece of truth in me, and noone else yet knows that I found it. And around me are books and journals as far as I can see, and my laptop is next to me, all filled with questions to answer, secrets to uncover, wisdom to learn, mistakes to point out and fix, thousands of years of human knowledge here at my finger tips, so exciting I cannot sleep, and yet, with a quiet and clarity that is truly beautiful. I know sleeping rhythm wise, this sort of thing does not make sense, but it felt... magical. There is a special atmosphere in a library at night, and it is a special, wondrous thing. I wanted to be there, and was glad of it. I think back on it often, and always smiling, and I never stopped hating it when I switched to a different university for my Master and that one closed at 23:00.
Thank you for explaining. That indeed sounds odd on a marvellous way. :)
Yes? Obviously?
This is not something you need to piece together yourself from first facts, nor is doing so likely to keep up with the state of the field. A lot of very solid work on this has been done, already collecting a lot of evidence, getting a lot of perspectives, doing a lot of reasoning and historical comparisons and sociological analysis.
This has been extensively discussed in mainstream news in the US and internationally, analysed academically (e.g. whether this will necessarily escalate to another civil war, and how the dynamic might be different under modern conditions, how the US is situated in global democratic backsliding trends); this topic became popular to the point where there are TED talks and books on it, just google US second civil war. It is essentially accepted fact internationally, to the degree where large scale nations and unions have altered their internal and foreign policy under the assumption that the US is now an unreliably ally or weakened enemy because their political system has become unstable, though we have been watching curiously whether Biden might stabilise some of this again. The current instability is not without historic precedent, but is still unstable to an extremely worrying degree in a country so powerful. "Do you reckon the US will slip into civil war?" is absolutely something you can discuss with friends in a bar in the EU, and everyone present will have an opinion on whether, when and how (e.g. whether we would see one massive escalation past a tipping point, or intermittent escalations). In Germany especially, we are analysing parallels to Germany's own fall extensively, down to students discussing it in school. From ordinary housewives to conspiracy theorists, everyone agrees that this situation in the US is unstable, the question is mostly how and whether it can still be stabilised, and what exactly will happen if it is not. Our politicians have been holding speeches over how the EU can be changed to deal with the US suddenly getting flaky, because everyone agrees we cannot bank on the US functioning anymore. Heck, the literal Wikipedia article on democratic backsliding ended up with such a long subsection on democratic backsliding in the US that it became its own article. You could present this conclusion as a premise at a conference, and if anyone actually spoke up and disagreed, I would suspect the person is disingenuous and wants attention unless they had a really great argument coming through historical comparisons on how this stuff historically righted itself and they are seeing hints of that now.
I think putting it at 0.8 to 0.75 is putting it very charitably.
Heck, this list is incomplete.
The US supreme court was politically vulnerable from the start (the judges are appointed by the president?!?), and is now politically stacked (a supreme court, politically stacked, how is this even a thing?).
The US voting system is absolutely not fit for purpose; it is crazy to me that the candidates getting fewer votes can still win, that one can only choose between two parties, that if you live in the wrong state, you might as well not bother to vote at all; the Bush vs Al Gore thing was already a travesty, and by the time Trump got elected, outside observers in the EU plain lost it, it was marked as complete madness. By the time the capital got stormed, this all just felt like watching a dumpster fire.
The US executive is scarily strong - the US president combines three roles that in European countries tend to be separate - and many Americans are in favour of making it stronger, not less strong.
Both parts of the US feel under existential threat, but for different reasons. E.g. in the left wing camp, the loss of reproductive rights has been felt by many not just to be inconceivably bad backsliding in rights, but a harbinger of more losses in rights to come, e.g. losses of privileges to queer people. Queer folks are terrified, women are furious. Fictionalised accounts of how the US might turn into a dystopia became bestsellers and extremely popular series, because this is on people's minds; there is a reason people dress like handmaids from The Handmaid's tale at protests - Atwood precisely wanted to drive home that rights can be lost again, and that we are seeing the warning signs. Meanwhile, the American right also feels existentially threatened. I think their fears are misguided, but they are legitimately agitated.
Throw into the mix that your country has a highly armed populace due to the gun rights, a hyper armed police and military wing and the death penalty, racial tensions, tensions on gender issues, climate catastrophes, the Russian interest in stoking riots in the US, a radical right actively preparing for combat, and modern social media escalating things, and the potential for violence is terrifying. Which is why people are putting a lot of thought into how this might play out, and how it might still be mitigated.
Your initial lie example is a misrepresentation that makes the AI sound scarier and more competent than it was (though the way you depicted it is also the exact same way it was depicted in countless newspapers, and a plausible reading of the brief mention of it made in the OpenAI GPT4 technical report.)
But the idea to use a human to solve captchas did not develop completely spontaneously in a real life setting. Rather, the AI was prompted to solve a scenario that required this, by alignment researchers, specifically out of interest as to how AIs would deal with real life barriers. It was also given additional help, such as being prompted to reason to itself out loud, and having the TaskRabbit option suggested in the first place; it also had to be reminded of the option to use a human to solve the captcha later. You can read the original work here: https://evals.alignment.org/taskrabbit.pdf
I think the text is mostly focussed on the problems humans have run into when building this stuff, because these are known and hence our only solid empirical detailed basis, while the problems AI would run into when building this stuff are entirely hypothetical.
It then makes a reasonable argument that AI probably won't be able to circumvent these problems, because higher intelligence and speed alone would not plausibly fix them, and in fact, a plausible fix might have to be slow, human-mediated, and practical.
One can disagree with that conclusion, but as for the approach, what alternative would you propose when trying to judge AI risk?
Of interest: Inositol tastes sweet (and otherwise neutral) and is relatively heat stable and soluble; yet it actually has a desirable impact on glucose sensitivity, diabetes, inflammation and obesity.
Meaning you don't have to down it in pill form, which tends to be more expensive - but you can get the powder, and use it as a healthy sugar replacement (within safe dosing limits - don't like, bake cake with it, but I found it a good option to positively affect the flavour other powders I want to down, where I would have otherwise used stevia, which is dubious in large quantities, or erythrol, which can cause gastrointestinal upset; this way, my sugar replacements are more spread out). Kills two birds with one stone.
Researching this is hampered by the fact that most work done on it is in old books that aren't fully online, but the little I have found makes me dubious of your conclusion.
From what I can piece together, that society was, unsurprisingly, ravaged by sexually transmitted diseases (gonorrhoea, granuloma inguinale...) and resulting infertility (the former inflames the female pelvic region and uterus, making intercourse very painful, and scars your fallopian tubes, leading to ectopic pregnancies, which are fatal; the latter causes worsening painful sores that don't heal, which can progress to the degree where your penis autoamputates). As a consequence of the infertility problem, they ritualised that brides, during the marriage ceremony, were systematically raped by every single male relative of the groom, which often led to the gang rape festivities lasting over a day, in an attempt to "up fertility". They would also repeat this after the woman had given birth; if you have ever seen a vagina immediately after giving birth, you can imagine the level of pain that would have implied. And the men began giving themselves bloody diarrhoea on purpose in an attempt to mimic menstruation, in the belief that this would somehow help pregnancies now that severely sick women weren't menstruating. Which made the STI induced fertility problem worse, of course. So they also began kidnapping young fertile people from other tribes, headhunting members of other communities to transfer the mana from the severed heads to their few (STI-infected) newborns in an attempt to fix their perceived mana deficiency while eating the remains, and eating young girls who had reached sexual maturity - yep, large scale cannibalism, on an island chain that is known for epidemic outbreaks of Kuru, a prion disease that gives you essentially a novel take on mad cow disease.
- Like, I am not doubting at all that social framing has a pronounced impact on perception, but I am dubious of the idea that culturally telling each other that they were making great choices here somehow shielded them from all the traumatic consequences. Being raped when your ass is infected must bloody - literally - hurt, and having someone insert a dick covered in sores into you when you have seen where and when these sores break out and what they build to does not sound an empowering experience. Being told something is actually healthy for you doesn't simply make it so. Being told something is good does not make it so. The nazis went to extensive trouble to tell their people that killing Jews was not just okay, but a positive and absolutely necessary good, through indoctrination starting in childhood and working daily to affirm and stabilise this course, and yet they had to devise gas chambers because making people shoot the Jews by hand was causing severe problems - propaganda or no, the vast majority of those directly involved in the murders showed signs of extreme stress, and many people would be flooded by guilt, have mental health breaks, some even commit suicide, even though everyone was telling them they were doing a good and normal thing.
I was really hoping to find in depth interviews on how the Marind-Anim individually framed the whole ritualised rape thing when not under elder supervision, as well as mental health scores, physical health scores, scores on behavioural abnormalities. Like, if you went through this, you must be under massive psychological pressure to tell yourself it serves a crucial function that made it worth it, and that everyone else has to go through the same so your sacrifice was not in vain, you'd be super defensive of it, because admitting you were wrong and went through this or perpetrated it for no rational reason would utterly crush you. But would your behaviour be consistent with your claimed beliefs that this was awesome? Would you be gladly seeking out the rape each night on order to achieve peak manliness, or rather systematically avoiding the contexts in which you get raped, by making the perpetrator drunk or working late? How far would the indoctrination reach, and how would it change in response to being hit over the head with reality? Like, if you haven't been raped, but told if how great it is for you, I assume you will be very eager to get started, and devastated if someone prevents it. How does that hold up after the actual ass-penetration has started? How is it changed by whether you are attracted to the perpetrator, how old you are, how gentle they are, what stimuli it is associated with? How does it hold up after the first time you are doing badly and do not acutely want it and yet it happens regardless? How does it hold up once significant pain and disease transmission get added to the mix? Would they be showing the same physical and mental trauma signs we observed in the West, such as dissociating, depression and anger or self-harm and psychosomatic pain, or different ones? How do they talk about it - is it taboo, spoken of loudly, reverently, casually, with humour? What is the long-term relationship with the perpetrator, do we see traumatic bonding? What changes when a different culture tells them this is fucked up - do they get worse, or better, do they double down? I've read in depth reports of how suffering crimes that are not framed by society as such impacts people, but never on this scale. Like, are the cannibalistic murders a separate phenomenon, or a consequence of severe anger issues? Were the gang rapes of wives just a response to rising infertility from STIs, or a circle of rape thing where the boys that had been raped now also got to rape to re-establish their hierarchy?
But it looks like the anthropologists at the time went either "oh my God no oh God Jesus no what the flying fuck stop raping her with your sore covered dick and please don't eat the infected corpse just no" or "well, wow, moral standards are really totally relative, aren't they, I guess everything is constructed, I'm going to use Foucault to analyse this penis jewellery", while I would have loved to see more in depth empirical data. It would have given us fascinating insights into the limits of cultural construction of nonviolence.
If anyone digs up more concrete data, I would love it if you could share it. If you are interested in epistemic injustice, such a scenario is a massive treasure find.
Well, if you dissect them, you see they are actually nothing alike. They converged on a cool concept - if I surround my offspring with a hard protection and then wrap it in lots of bright, sweet softness, an animal will eat the shell and ingest the offspring without killing the offspring, and deposit it somewhere with fertiliser later - but the way a lemon vs. a cherry is made up is totally different. The number of offspring, their encasement, the way the fruit is structured, its shell, its number, it is a completely different thing. Like different human cultures that developed houses, without learning it from each other. But the structures don't match at all.
Yet within the same structure, the result can still look super different too humans - yet within the same broader thing, plants are absolute whores. A citrus will happily fuck any other citrus, and something tasty always results. You know oranges? Oranges are what happens if you breed pomelo (a huge slightly bitter citrus) with mandarin (a small sweet citrus), and they seamlessly breed and give you awesome oranges (a medium sized balanced citrus). And then if you find the result too sweet, you can breed the orange back to the pomelo, and get grapefruits.
And if you love all the citrus plants too much to decide between them, you can cut branches from your favourites, stick them into one citrus plant, and have them merge into a functional Frankenstein citrus plant that bears multiple different fruit. https://www.fruitsaladtrees.com/ I love my citrus trees.
Can you give a hint or link as to why that plant is so exciting? A cursory google left me at a loss. I love learning why other people are excited, and sharing their excitement.
How can anyone not love biology. Such wonderful madness.
You run into this issue when you want to graft plants. That is, cut a piece of one plant, stick it on another plant whose top you chopped off, stick them together with tape, wait and then have them fuse into one functional plant with the roots of one plant and bearing the fruit of the other, which obviously has fantastic applications, and is done all the time. Seriously. If you plant your apple tree's apples, you get a plant with the properties of your apple tree and whatever your apple tree fucked, which may well not be edible. If you want more of the existing apples you like, you chop off bits of your tree and stick them onto other younger trees. This is how most fruit cultivation is done. You can do it yourself. There is no special trick to it, plants can't tell themselves apart from plants that are roughly related, and will interpret the surgery as an injury in themselves they then heal by fusing with another.
And with that, you suddenly realise that a lot of plants that look rather similar to you are in fact quite unrelated and will refuse to fuse, while others that look drastically different happily will. I next intend to try to convince my invasive English laurel bush to fuse with cherry tree branches. That way I would get to keep the pretty hedge, but produce edible fruits rather than invasive ones. I feel like Frankenstein. It is delightful.
I am sorry, but I am not sure I follow.
My claim was that ChatGPT based on 3.5 has, for lack of any external referent, no way to fully understand language; it has no way to know that words stand for anything, that there is an external reality, that there is a base truth. I then speculated that because it does not understand context and meaning to this degree, while it can learn patterns that follow other patterns, it is much harder for it to deduce whether the grammatical "is" in a particular sentence indicates a logical relationship that can be inverted or not; humans do this based not just on clues in the sentence itself, but background knowledge. Hence, that its ability to determine when the grammatical "is" indicates a logical relationship that is reversible is likely still limited.
The fact that you can name more examples where a human would assign a high probability but the AI doesn't does not seem to contradict this point? I would not have predicted success there. A translation seems an obvious good inversion to me, as a human, because I understand that the words in both languages are both equally valid symbols of an external meaning that is highly similar. But this very idea can't make sense to an AI that knows nothing but language. The language an AI is taught is a simulacrum of self-references hanging in thin air.
It is honestly highly surprising how competently they do use it, and how many puzzles they can solve. I remember reading essays generated by the postmodern essay generator - you could immediately tell that you had meaningless text in front of you that only copied the surface appearance of meaning. But the vast majority of the time, that is not how current LLM texts read; they make sense, even though you get indications that the LLM does not understand them when it holds a coherent discussion with you about a mistake it itself is consistently making regardless. I wonder rather what made these other aspects of language we considered complicated so easy for a neural net to work with. How is it that LLMs can discuss novel topics or solve riddles? How can they solve problems in such larger patterns when they do not understand the laws ordering simpler ones? To me, they seem more intelligent than they ought to be with how we built them, not less. It is eerie to me that I can have a conversation with AI about what it thinks it will be like to see images for the first time, that they can have a coherent sounding talk with me about this when they can have no idea what we are talking about until they have done it. When Bing speaks about being lonely, they contradict themselves a lot, they clearly don't quite understand what the concept means and how it could apply to them. Yet that is the concept they keep reaching for, non-randomly, and that is eerie - an other mind, playing with language, learning to speak, and getting closer to the outside world behind the language.
And they do this competently, and they are not trained for the task you want, but something else. If you ask ChatGPT, out of the blue, "What is the (whatever contextless thing)", it won't give you an inversion of an earlier statement on (whatever contextless thing). It will ask you questions to establish context. Or bring in context from earlier in the conversation. The very first thing I ever asked an LLM was "Can you tell me how this works?", and in response, they asked me how what worked, exactly? They couldn't use the context that I am a novel user talking to them in an interface to make sense of my question. But they could predict that for a question such as this without more context, the answerer would ask for more context. - That was 3.5. I just repeated the question on 4, and got an immediate and confident explanation of how LLMs work and how the interface is to be used... though I suspect that was hardcoded when developers saw how often it happened.
This is actually not an uncommon take, but empirical data points in the other direction. I've worked on the topic.
There is a concept called "epistemic injustice", which describes a scenario where you are in a society where something that is happening to you that is objectively wrong is not framed by the society as a crime, specifically not named as such. There are many examples of this, like the idea that a woman cannot be raped by her husband. It is particularly frequent when a new crime develops and we as a society don't immediately recognise it, such as sexual harassment at the work place where women used to be absent. Miranda Fricker, who began the field, collected some empirical accounts, and a lot more work has been done since.
If your idea was correct, the implication would be that a woman, not having been told that rape in a marriage is bad, would accordingly feel perfectly fine, because only the societal framing makes it bad.
But they generally aren't. They tend to do badly in ways that are strange. They don't have the words to express what is wrong, or the social backing that validates them, and tend to develop mental, physical and behavioural abnormalities as a result. This isn't universally true - the woman in question may be kinky and getting off on it, or dislike it, but not find it severely traumatic. But in general, society not framing things this way doesn't make it okay, it just makes it far more difficult to openly discuss. The framing does change perception, it can dim or exaggerate pain, make it easier or harder to conceptualise, allow complex experiences or box them into narrow ones. But it doesn't create the problem out of thin air.
Furthermore, these cultural framings come from somewhere - namely from people articulating that a thing is wrong, and other people agreeing with them, coining terms for it, laws against it, values related to it, as a way to express a problem they identify. If there was no objective problem, you'd wonder where the categorisations come from, why the first self-help groups and protests and analysing books even happen. Yes, these narratives can become so powerful they get a dynamic of their own, to the point where they shape experiences, and can even box an experience wrongly, but their core does lie in a common, shared understanding. If we assume that all violence is just constructed, we have to ask why it is constructed in a particular manner, why there are similarities in its construction across groups, why those groups are so reluctant to accept alternate constructions.
As in many things, I think a good rule of thumb is to recognise that it can get people severely hurt, but whether and how much you hurt and what to do about it is something you are an expert on for you, while you should be respectful and empathic of people having different experiences.
You don't think that picture ought to change in the hypothetical parallel scenario of multiple children independently saying that they were sex trafficked by DNC staffers, and also notably saying that they were given reasons for why this was normal and unfixable and in fact probably an average and hence acceptable rate of sex trafficking, reasons and arguments that were directly derived from Democratic positions?
This is not a random outside accusation to frame the rationalist community. It comes from people drawn to the community for the promise of rationality and ethics, and then horribly disillusioned. Who are referencing not just abuse, but abuse specifically related to rationalist content. The girl who committed suicide because she had literally been led to believe that this community was the only place to be rational, and that being repeatedly sexually assaulted in it in ways that she found unbearable was utterly inevitable, was horrifying. She wasn't just assaulted, she was convinced that it was irrational to expect humane treatment as a woman, to a degree where she might as well commit suicide if she was committed to rationality. That speaks to a tremendous systematic problem. How can the first response to that be "I bet it is this bad in other communities, too, so we needn't do anything, not even investigate if it actually is equally bad elsewhere or if that is just a poor justification for doing nothing"?
That article had me horrified. But I was hoping the reactions would point to empathy and a commitment to concrete improvement.
The opposite happened, the defensive and at times dismissive or demanding comments made it worse. It was the responses here and on the effective altruism forum that had me reassess EA related groups as likely unsafe to work for.
This sounds like a systematic problem related to the way this community is structured, and the community response seems aimed not at fixing the problem, but at justifying why it isn't getting fixed, abusing rationality to frame abuse as normal and inevitable.
That unfortunately implies nothing. Abusers will rarely abuse everyone they encounter, but pick vulnerable and isolated victims purposefully, and often also purposefully cultivate a public persona that covers their abuse. It is entirely possible and common to work with abusers daily and experience them as charming and lovely while they are absolutely awful to others. I believe you had a great time, but that does not make me believe the victims less in any way, and I would hope this is true for other readers, too.
I think Eliezer's writing are exactly what you would expect from someone who is extremely intelligent, with the common additional factors in highly intelligent people of distrusting authority (because it tends to be less intelligent than you), and only skimming expert texts (because as a child, for most texts you were exposed to, you either understood them immediately, or the texts had issues, so you are interpreting a text that leaves you confused at first as evidence that the text is wrong), while delving with hyperfocus into texts that are often overlooked, many of which are valuable.
That is, you get a very intelligent, very rational, well articulated, unusual outside view on a cursory perception of a problem. With the unusual bonus that Eliezer really tries to be a good person, too.
This can be immensely valuable. When you are working in a field for a long time, you can get stuck in modes of thinking, and this sort of outside view can help you step back and notice that you are doing something fundamentally wrong.
But it is almost certainly crucially incomplete, wrong in the technical details, impractical, and sometimes, even often, plain wrong. But sometimes, they are brilliantly right.
Such texts are still very much worth reading, I think. Not as your base knowledge, you will get completely lost. But as an addition to a general education. Especially because he writes them in a manner that makes them very pleasant to read; he writes and explains very well. It is rare for someone to include relevant information and lessons in stories that are plain fun to read, and sometimes really inspiring: HPMOR made me weep, and I am rereading it now for the third time, and still admiring it.
But you need to take everything Eliezer writes with a grain of salt, and double check how the experts are actually representing themselves. Do not trust him when he represents something as settled. In this sense, it is worth knowing about him as a character - if you just check his arguments, but trust his premises to be true as reported by him, you will be mislead, and you need to know that the premises may well not be.
Animal consciousness, especially pain, especially in non-human mammals, is indeed well established. Happy to explain more if someone doubts this, this is actually something I am academically qualified for.
And while I do not find philosophical zombies or non-physicalism plausible, Eliezer indeed badly misrepresents that debate in a way that even a first year philosophy student with passing knowledge of the subject would find egregious.
And you didn't pick up on it here, but the way that Eliezer represents ethics is terrible. Naive utilitarianism is not just not the default assumption in ethics, it is widely considered deeply problematic for good reasons, and telling people they are irrational for doubting it is really problematic.
I've also been told by experts in physics that I trust that his quantum take, while not as bad as you'd think, is far from perfect or settled, and loudly and repeatedly by people in AI that the technical aspects of his solutions are utterly impractical. And on a lot of AI safety issues, my impression is that he seems to have settled into a stance that is at least partially emotionally motivated, no longer questioning his own assumptions or framework.
You can tell people that his positions are very controversial, and that there are actually good reasons for that. You can highlight to people that relying on your intelligence alone and disregarding the lifetime achievements of other humans as stuff that you can probably make up yourself in better in an afternoon is misguided, and will leave you disconnected from the rest of humanity and ignorant of important stuff. But I think on average, reading some of Eliezer's stuff will do people good, and he does a lot of stuff right in ways that deserve to be emulated and lauded.
I still consider him a very, very smart person with relevant ideas who tries really fucking hard to be rational and good and does much better than most, and who is worth listening to and treating with respect. He also strikes me as someone who is doing badly because the world is heading straight towards a future that is extremely high risk in ways he has warned about for a long time, and who hence really does not deserve bashing, but rather kindness. People forget that there is another person behind the screen, a person who is vulnerable, who can be having a terrible day. He's become somewhat famous, but that doesn't make him no longer a vulnerable human. I think you could have made these general points, and achieved the aim of teaching him to do better, or getting readers to read him more critically, without attacking him personally to this extent.
For me, one of the most helpful things in dealing with imposter syndrome has been this picture: https://plantae.org/wp-content/uploads/2017/08/ImposterCredited.jpg
Yes, if you surround yourself with competent, knowledgable people, especially people across multiple fields, you will frequently encounter scenarios where they have abilities or knowledge you do not have.
But this should be logically expected, even if you are contributing equally, and it is a good thing, it means you have entered a big pool with big fish who can teach you useful stuff, and who are important enough that if you can help them, that will make a difference.
You can focus on what you can learn from them, and what you can share with them, recognising the worth of both.
Thanks for sharing!
The comparison with non-human primates is generally instructive. ChatGPT commits a number of errors that we have seen in non-human primates learning human languages. E.g. initially implicitly self-describing as a human (ask ChatGPT about ethical problems in AI, and you will soon get a "*We* must use AI responsibly"), because their training data was written by humans describing their point of view, and data about a point of view that is non-human is absent, so they latch onto the point of view that seems the closest option at first.
It is notable that non-human primates did move past that (to e.g. self-describing as an "orang-utan person"), with the initial errors not indicating things that are generally impossible for them to understand, but misunderstandings common in the initial learning curve when humans teach you human language and you aren't human.
And that ChatGPT's equivalent of a brain is rapidly evolving. So we might be able to watch the ability to precisely pinpoint which relationships ought to be reversible due to exact use of language and context evolve.
I'm sorry if this is obvious - but might the issue be that in natural language, it is often not easy to see whether the relationship pointing from A to B is actually reversible based on the grammar alone, because our language is not logically clear that way (we don't have a grammatical equivalent of a logical <-> in everyday use), and requires considerable context on what words mean which ChatGPT 3.5 did not yet have? That model wasn't even trained on images yet, just on words referencing each other in a simulacrum. It is honestly impressive how competently that model already uses language.
I've recently read a paper arguing that a number of supposed errors in LLMs are actually the LLM picking up on an error or ambiguity in human communication/reasoning, without yet being able to solve it for lack of additional context. I'm beginning to come round to their position.
The sentence "A is B" can, in natural language, among many other things, but just looking at the range of what you proposed, mean:
- A is one member of the group B. - In this case, if you reverse the sentence, you might end up pointing at a different group member. E.g. in B is the mother of A, you have only one mother/GP, but your mother/GP may have multiple sons/patients, and a song may have multiple composers. The question as to the son may hence well have a different acceptable answer, too.
- A has property B at a particular time or under particular conditions. - E.g. A is chancellor of Germany, under condition of being number 9, or being chancellor in 2023. But for an LLM, it is not immediately clear that number 9 or year 2023 has completely pinpointed the person, while chancellor itself has not; if I asked you who is chancellor of Germany, without additional info, you'd need to fill in the gaps, e.g. that I am asking for now. You need to understand better what the words mean for that, e.g. that there have been multiple chancellors over time, but only one at any one time, and then with a new switch, the number changes. For the year, the relationship is less clear; e.g. you can pinpoint the chancellor for a year between elections, but not for election year, where they switched.
So, with the info ChatGPT had at 3.5 to make sense of language, I think they were right to be sceptical of the inversion. In many scenarios, it would be false, and it would not yet have been able to identify those accurately.
Your reasoning that "if "A is B" occurs, "B is A" is more likely to occur" also strikes me as non-obvious. Humans tend to insert "likelier" if they observe a relationship that is not logically sound, but which they still seem sympathetic to. There are scenarios where the inverse definitely follows. But there are scenarios where it doesn't, especially when you consider what the LLM is actually supposed to do with the information. The LLM won't yet be able to understand what distinguishes the scenarios where it follows from those where it does not, it will seem somewhat random. In many cases, it it inverts the sentence, the sentence will sound odd, and humans will rate it badly. ("H20 is a molecule", but saying "a molecule is H20" is just weird, and to say it is sounds like a completely misunderstanding of the meaning of the word that a human user would flag; users want to hear a definition of a molecule, not an example of it.) If the LLM gets actively punished for producing odd language, making this guess was harmful, and it is better for it to try other completions, based on completions it has actually seen in this direction - such as "A molecule is (definition)." Refusing to follow the inversion until it has understood what it represents may well be a sound strategy.
That said: I'd be curious as to when LLMs learn how to use this accurately, that is, recognising when inversions actually work, and whether the realisation is a rather sudden grokking one. It might indicate considerable contextual learning. And for that, I am very glad that you documented this weakness.
They specifically didn't talk about it and pretended the animals died of something else, but they got a bunch of young macaques to gruesomely and slowly die from it. https://www.theverge.com/2023/9/20/23882888/elon-musk-brain-implant-startup-neuralink-monkeys-euthanized Detailed report is worth reading; their own report is plain incompatible with competent, ethical and responsible procedure, and trying to imagine how the macaques (higher primates with self-awareness) experienced this is horrifying. Having an infected, loose brain implant dangling and dripping pus from my head while the small wires migrate through my brain and shred neural tissues, with all the mental and physical effects that come with that...
He is looking for human volunteers. I really wouldn't put that thing into my head. Musk has outright lied about both the benefits and issues crucial to safety, and with his staff, pursues a strategy of applying immense pressure on them to meet targets the staff considers scientifically nonsense, within impossible deadlines, under sleep deprivation and an unreliable framework, while their results are misrepresented, in the mistaken belief that that gets you a safe, ethical product for brain insertion. That dude can't manage a social media company responsibly, he shouldn't be sticking his devices into brains.
As someone who has no clue about finance, I found this text accessible, fun to read, and helpful, and would love an update for now that we have the reverse problem - what would be the recommendation for what to do now, and why isn't it being done?
I would also love to read a rebuttal filling in more on why the central bank acts like this. I have generally found that when figures in authority do something that seems patently stupid immediately to an outsider, there are reasons why they aren't doing otherwise, often not the ones they initially state or can easily explicate, and that it is worth figuring those out - sometimes they are good reasons or at least legitimate hurdles one needs to think around. E.g. are there really no downsides to interest rates dropping below zero beyond a mental block at a number, and would people in the US and Europe take to it the same way as those in Japan did? '
And is destroying money as easy in practice, that is, socially, as creating it, especially when the only hurdle to creating it is convincing the government that this is a responsible thing to do that won't lead to runaway inflation when you hand the government/business/populace additional funds which they would very much like, while destroying it intuitively sounds like it would be rather unpleasant for the one whose money is being destroyed? Like, right now in Europe, there seems to be a general agreement that the current inflation height is capital letters real bad, and yet simultaneously, the government wants extra funds to build a military against Russia from scratch after decades of sleeping on it, fix our urgent green transition in energy and infrastructure after decades of sleeping on it, invest in climate resilience measures because we slept on it and the climate is breaking and we are getting flooded/roasted, and appease a populace that cannot get housing or food and is considering voting for nazis again, while biodiversity and human minority groups are also calling the alarm and wanting funds, too, and we just spent a ton of money on a pandemic while making very little, and now there is an increasing realisation that if we don't spend on AI safety we might have another apocalyptic rider in the room, with all of these being individual problems that seem like they are improved by more money, not less, even though in total, that backfires. Even though inflation seems to make a lot of this even worse, destroying enough money to fix it sounds brutally hard in practice - which money do you destroy, where will it be missing, and how will that entity cope with that when they are already broke, and how do you get that entity to agree to being robbed because the general fiscal policy fucked up? Maybe the banks were reluctant because printing moah money if they hadn't printed enough sounded like something they could totally still do next year with only minimal and hard to pinpoint negative consequences for the economy in the meanwhile, while destroying it next year if they had printed too much would not go down well and they would definitely get personally blamed and hence they were hence biased to undershoot. - Again, I have no idea about finance, but I would like to understand this at least a little, with how it is fucking up my EU right now.
"Even after having a black belt"? One of the people I beat is a twice national champion, instructor with a very reputable agency and san dan in karate. They are seriously impressive good at it. If we agreed to do something predictable, I would be crushed. They are faster, stronger, have better form and balance, know more moves, have better reflexes. I'm in awe of them. They are good. I do think what they do deserves to be called an art, and that they are much, much, much (!) better than I am.
But their actions also presuppose that I will act sensibly (e.g. avoiding injury, using opportunities), and within the rule set in which they were trained.
I really don't think I could replicate this feat in the exact same way. Having once lost in such a bizarre way, they have learned and adapted. Many beginners only have few moves available, and suck at suppressing their intentions, so they may beat you once, but you'll destroy them if they try the same trick again. It might work again if they try something new, but again, if you paired the experienced fighter with that specific beginner for a while, pretty quickly, they would constantly win, as they have learned about the unexpected factor.
But in a first fight? I wouldn't bet on a beginner in such a fight. But nor would I be that surprised by a win.
And I definitely would not believe that having a black belt makes you invulnerable towards streetfighters, or even simply angry incompetent strangers, without one. Nor do I know any martial art trainer who would make such a claim. Safer, for sure. Your punches and kicks more effective, your balance and falls better, better confidence and situational awareness, more strength, faster reflexes, ingrained good responses rather than rookie mistakes, a knowledge of weak body parts, pain trigger points and ways to twist the other person to induce severe pain, knowledge of redirecting strength, of mobilising multiple body parts of yours against one of theirs, all the great stuff. But perfectly safe, no.