That… is a good point

It obviously did matter when same-sex marriages were not legal for example (some countries still do this…), but in most parts of the developed world today… yeah. I guess it does feel a bit grandfatherer-in right now

Obviously things such as the medical system need to know what biological sex someone is, but they don’t have to tie into the legal system

Wanna try hugo? I can’t vouch for the others, but since I have tried it I can share my experiences… It’s not 100% as intuitive as Microsoft Word so there will be a learning curve, but it is reasonably simple & can scale up to professional web development quite well. Lots of templates out there too so for the most part you won’t need to make everything yourself. Some professional web devs use it too. But because it is also used professionally, AI is good at generating Hugo sites, so it may make your site look AI-my…

Just a heads up though, most hugo themes use JavaScript quite heavily; if you’re against JS by chance, you may need to dig a bit deeper into their themes

My trade, bioinformatics/computational biology… a ton of software documentation, tutorials, guides, vignettes… And some programming books here-and-there (which the authors released online for free). There are Master’s programs, but realistically hands-on researchers overwhelmingly learn by doing. The field is just too small and scattered… The downside of this approach is that my field is traditionally criticized for having little to no adherence to software engineering best-practices

My hobby, there are like 2 websites (one is a dedicated webpage, one a GitHub wiki) that contain literally everything one needs to know about them

I mean that is a higher pay package than what I got as a PhD student ($16-18/hr before tax, nearly no benefits) so there was no way I’d have said no to that… At this point no, but it’s only because my total compensation is a lot higher, and I could have afforded such a living arrangement and still have more leftover money

If the compensation is comparable, I probably wouldn’t mind either way. But then I don’t have a family so…

Fun fact, there are actually a few NYC-based PhD programs which provide heavily subsidized student housing (mostly on Roosevelt Island I believe), conditions are somewhat parallel to what you proposed

Not really lol, both Japanese and Chinese heavily use standard 26-set alphabets nowadays

Japanese alphabets (Hiragana/Katakana) have their alphabet counterparts. Aside from L/R which is confusing (the R sound doesn’t exist in Japanese, and L is written as R), almost everything else intuitively makes sense. Chinese is almost always spelled out in Pinyin (which is in standard alphabets) for really young kids, and ppl predominantly use alphabetic keyboards for typing. If anything, the reverse is kind-of true in China, where kids are having struggle remembering how to write proper Chinese characters…

The weird pronunciations and grammar rules are probably more difficult IMO. But then difficult pronunciations and grammar rules exist in other Germanic/Romance languages too (try pronouncing a G in standard Dutch or an R in French or German!)

• How come the most memory-inefficient ML model we have ended up being the most powerful one. Can we rewrite the universe so modern advanced AI is based on network inference or ensemble models or sth
• How come AI has taken over the economy, and it doesn’t seem like a bubble (or at least not the crypto type of bubble)
• I’m a bit surprised that the most predominant use of AI/LLM seem to be… as a search engine wrapper. Thankfully ChatGPT is getting better at it, their default model has a built-in search engine integration now. This is kind of important since the company is getting big enough that their model bears social responsibilities…
• I have an asshole-ish work environment where the head honchos only cares about professionalism and efficiency… so I have no shame using AI for work. They don’t deserve my best tbh
• In terms of actual usage for my work, AI outputs can be anywhere from very good to very bad in an instant, and sometimes it’s hard to tell the two apart…
• It’s surprisingly helpful as a primer to help me with things I’m not good at; researching niche information, finding resources
• For personal usage, I’m still hoping that there is a way to work AI into some type of a home automation loop… I haven’t found that sweet spot yet. Maybe it’s just my lack of skill at the moment. Hopefully I can hook together some shitty models with search engine skills (on my own SearXNG) as a semi-useful agent

Originally from China. It’s probably quite hard to explain on a short post over the internet especially since there are way too many fucked up things where I’m from, but I absolutely wanted to and have been living outside for 11 years straight now, and I’ve been mostly happy wherever I’ve been (and I’ve been in some shittier cities). Although it is more of an “anywhere but China” attitude for me…

So yeah. I don’t think I necessarily dreamed of living in any particular country, but just wanted to live anywhere else

I’ll go one by one… although I do have to say, I can’t answer the main question of whether it can be 100% ethical

Open-weight model

It’s actually quite literal. LLMs may look like magic, but under the hood are basically very, very big mathematical/statistical functions that transform one type of input (for example, “User: hey ChatGPT what is an open-weight model?”) to an output (example, “ChatGPT: To understand what an open model is…”)… To simplify, these models function by piecing together a ridiculous amount of tiny, nonlinear mathematical functions, apply a weight to each of these tiny functions, and somehow get a coherent result out of it. The theory behind this has been known in math research for decades before OpenAI.

Because LLMs are mathematical models, if you know the weights of those tiny functions and how you chain theme together, you can run an exact replica of it. Open-weight models are just that, models where the entire weight and model architecture are released with open, permissive licenses. Many of these end up being modified by other users/hobbyists, and I do believe this sharing/modifying part of AI/LLM is quite true to FOSS philosophy.

As others have mentioned though, “open-weight” has no bearing on how the model is trained. It just means that the final product is free and open

Is it really feasible to run it 100% locally?

Theoretically yes, practically yes but depends on the model.

• For the really small models designed for mobile phones, absolutely. For example, LFM-2.5 and Ling-3.0-tiny run on just about any potato. Granted these models are… not great. They are probably functional as “search engine wrappers” though, methinks that’s what they were built for.
• For the intermediate models, yes but with an asterisk. They require at least a decent GPU, so either you have a gaming rig, happened to have snagged one of the AI computers, or try to kneecap the model in some way (limit the amount it generates, make the mathematical operations less precise a.k.a. quantization, be okay with it running for a long time, …). I have a local setup for Qwen3.8-27B with 4-bit quantization, but it runs at 2 tokens/second on my laptop (which has a ton of RAM but not a very good GPU)…
• For really large models, you can technically run them, but they are realistically designed for organizations with privacy in mind: hospitals, private companies that handle sensitive data, etc…

The software doesn’t come from nowhere

I mean IMO that’s the heart of the whole debate. Making a modern LLM requires massive amounts of data and computation… both of these do have to come from somewhere. And my understanding is that, current AI/LLM research didn’t create that much technological breakthrough, so building a better model often does indeed rely on either using more data or bigger architecture (which takes longer to compute)… which then leads to a truckload of ethics issues here and there

For private companies, we know that OpenAI is blatantly hoarding data from just about everywhere, see the recent Mathematics Millennium Prize scandal. A lot of the open-weight models are trained by Chinese training houses, and Chinese tech companies are not known for respecting intellectual property… and China uses lots of coal energy still.

Maybe Mistral is more ethical, given that 1) they are based in France and care about GDPR and data regulations a lot more, and 2) France uses lots of nuclear energy, but I can’t say for sure. An European company operating an AI/LLM training house from Iceland using their excess geothermal energy is probably the most ethical we can get on this

In theory (and take this with a whole jug of salt) someone can run an off-grid farm and train a small LLM all on their own, which would be very ethical… but I doubt any serious organizations are doing it this way. Most ppl just want better performance and/or more optimized models for now

For a complete FOSS purist with an environmentally conscious angle, I don’t know if there is an AI/LLM that can be considered 100% ethical… but then again Gentoo is a thing. So yeah, I don’t know

If my understanding is accurate… mostly healthcare accessibility

Japanese foods aren’t as healthy as most ppl would imagine, there are lots of fried stuff and high-salt stuff. And wealthier subsections of American cities can have life expectancy values hit close to 90 (see this Chicago map for example, the Loop has sub-parts where I was told life expectancy goes 90+)…

Also a side note, the 100+ observation might also be biased by bad data collection. See this primer/summary by The Guardian. The “Blue Zones” in Japan which have tons of super-centenarians tend to be in poorer areas which were bombed during WW2 and lost birth records

Chinese native speaker, pretty good at Japanese (passed N1 many years ago). I actually think spoken Mandarin Chinese is a lot easier to learn for most situations except for a complete beginner:

• For a complete beginner, Japanese has one of the (if not the) fewest possible valid pronunciation among all languages. There are literally only ~50 “alphabets” (officially only 48, there are a few weirder non-official ones), all of which are easily pronounceable. Mandarin Chinese also has a pretty restrictive pronunciation set, it’s simply that the tones (especially the 3rd) tend to trip ppl up
• Past the complete beginner stage and Japanese becomes noticeably more difficult than Chinese. Japanese grammar is an absolute nightmare, and you do have to know most of the grammar rules; not to mention you can have multiple different ways to pronounce the same word… Whereas Chinese has almost no grammar and overwhelmingly relies on memorizing vocabs, which someone who only wants to speak doesn’t even need to worry about how to write them (which is most of what makes it difficult)
• Just to clarify, both Chinese and Japanese are high-context languages (to compensate for way too many words sounding exactly the same), so the fact that a local might not understand you often has less to do with shitty pronunciation, and more to do with that the entire sentence makes no sense. A lot of Chinese ppl understand regional dialects (which are pronounced very differently) just fine

The Chinese dubbed anime thing is for Chinese ppl too. JP has a very robust seiryu (voice actors) industry, and a lot of CN/KR dubs just simply can’t catch up with the voice-over quality… A good example, think Genshin Impact: it’s a Chinese game but the default VA is JP (edit: to correct myself, it’s the de facto VA most ppl use; a lot of Chinese ppl happily switch to JP VA whenever they can). I highly doubt this has anything to do with the language themselves. The language themselves… I’ve heard some ppl say that spoken Chinese sounds “angry”, might have something to do with the tones. Not sure about Japanese…

Personal domains with custom emails. For email provider I use mailbox.org, which does allow custom domain mapping. The domain and email provider combo won’t be free but shouldn’t cost a lot

It’s a lot easier to deal with corporate spams if all of them are emailed to, say, advertisements@example.com or shopping@example.com, or something more on the nose. Paired with a password manager and it is possible to tell exactly which service is leaked, if I’m neurotic enough to assign randomly generated email names

Way too many for Chinese (think every Chinese dish you know), but the one world I think about the most in this context these days is 老百姓 (lao3 bai3 xing4) (Also see a somewhat related Wikipedia article). Direct translation is “ol’ hundred surnames”

It’s… a non-formal word for the “commoners” or the “average person”. This is a concept I don’t hear in mainstream media that much nowadays. I don’t think most news or mainstream media even refers to say the entirety of the lower to middle social class in one word… (Note that it is different from “Average Joe”, which does have a Chinese counterpart which translates to “Zhang three Li four”)

On top of it, the whole concept of “a hundred surnames” representing most of society is kind of absurd in English-speaking countries. Societies with historical Chinese influences tend to have lots of ppl share the same surnames

I think it’s close? Like 5-20 years close for commercially available robotics prosthetics systems (probably still need doctor’s prescriptions for it though)

• We already have pretty well-established prosthetics systems at the moment
• China and Japan (esp China) have some pretty exciting robotics breakthroughs, see the hilarious robotics Olympics from China recently
• However, China isn’t the best at manufacturing QC yet, so widespread deployment may take a few years from above

First of all: you’re good! I think there are lots of centrist and right-leaning people in academia who are doing just fine; especially in the US lots of world-class unis are in deeply conservative areas too. Most of academia don’t discuss politics that much at all as long as one cares about science. I’d wager the political leaning bias is more from a population bias rather than systematic discrimination of any sort

I could imagine AI researchers being more right-leaning yeah. The stats I got were from personal anecdote and research being done about this a short while back… there might have been an update in the AI age

First of all, good for you lol

It’s € 1.715 with today’s exchange rate. The minimum salary where I live is roughly € 1,9k after tax. Not life-changing money, but it’s still an absolutely huge amount that is enough for survival & is quite significant

I mean this is a large amount of money even for Americans though. The absolute highest paying “normal” job I could think of is a stacked SF Bay tech guy who makes $200K/yr after tax… but that’s still like 12% of take-home pay

Current STEM researcher who also happened to have met a few academics on the right side of the political spectrum… I am very qualified to speak about this I hope. I’ll try to use as unbiased language as possible

First of all, you are correct. It is generally true that most academics are left-leaning (obligatory not all, but you get the point)

In general, higher education attainment is associated with being left-leaning: academia is known to have extreme levels of education attainment (many jobs require a PhD). On top of it, academic jobs reward intellectual curiosity, do not tend to pay super well, tend to heavily rely on governmental funding… all of these are traditionally more associated with the political left or left-leaning. Still, there very much are right-leaning academics, they just don’t tend to scream around a lot about it

There are notable exceptions however. Economics is quite famous for being politically balanced, although what economics treat as left/right-leaning might be a bit different from the general public. If I remember, business schools (yes there are business professors!) tend to be more right-leaning. This is a smaller field and a bit more anecdotal, but quite a few prominent meta-science researchers are right-wing nutjobs: the “Why Most Published Research Findings Are False” guy and the NIH director Bhattacharya guy are both COVID deniers

There are also fringe scientists as some others have mentioned, but most people within their fields don’t consider fringe scientists as part of the academia so to speak

The Chinese side of the story is heavily censored and only has official reported numbers (likely tampered with, knowing China) and some propaganda pieces, but this is very much a Chinese national emergency as well. The Guardian’s latest reporting claimed that the Chinese officials reported 16 deaths and 546 missing, while Nepal’s numbers are 675 deaths and 2,498 missing

Maybe there are ulterior motives, but I think given the extent of the disaster, it is much more likely that China treated it as a top-priority national emergency, so might as well help out the neighbors with any extra manpower as well

My understanding is… Being popular (a.k.a. having lots of subscribers) requires someone to be entertaining and be good at self-promotion. Being entertaining is independent from being a good person. For the latter, I once saw a streamer say something along the lines that anyone who is willing to do that level of self-promotion likely self-selects for at least some level of narcissistic traits

… and a lot of their subscribers are relatively young and likely don’t care, probably

The idiom is “reduce, reuse, recycle” in that order for a reason. Even if plastic recycling is difficult in the US, reducing and reusing are doable by just about anyone. This can be using less one-time plastics, eating less meat (make sure nutritionally balanced), drive less, reuse those shipping boxes for other stuff such as for paper waste, buying from local farmers (if you can afford it)… something along those lines

For home, living in an apartment is actually one of the best things one can do for the climate, since living in an apartment in a city is much, much more energy efficient than living in a suburb… not to mention cutting down on car-related costs. For ppl living in a home, methinks AC is more of a boogeyman… having good home insulation is a much better solution (as a well-insulated home uses less energy!), and the best part is that it is bipartisan so red states would happily do it too. If renovation is not possible, stuff like blackout curtains, window blinds, … are cheap insulation options

IMO these are still negligible effects on an individual level, but these are not nothing, and if everyone does these the world would be a much better place. For more meaningful actions, probably organizing

Slay the Spire 2. I do NOT want to be infinitely killed, memory-wiped, and resurrected