Ask HN: What's the most important problem in the world? Are you working on it?
Everyone has different perspectives and values, so what do you personally think is the most important problem in the world right now?
The original quote is from Aaron Swartz [1]
> What is the most important thing you could be working on in the world right now? ... And if you're not working on that, why aren't you?
[1] https://www.azquotes.com/quote/1202481
90 comments
[ 37.8 ms ] story [ 187 ms ] threadthey are man's best friend, therefore anything else should take a back seat
So that the human society can move through the Great Filter and become a galactic species eventually.
- We cannot rely on any system. We need to rely on a human race that is not afraid to manage itself, so that we can mitigate the influence of power hungry people;
- So that we should force people to manage, even if they don't enjoy it. Because only by having the power to manage the state the people can retain other powers. If we leave management to a group of elites eventually those power will be taken away;
- This also means everyone should have a great amount of art and scientific knowledge, as management is both a science and and art these days. Public education these days doesn't really work very well;
- All management positions should be rotated if possible. This is probably too ideal but once most people reach a higher degree of self-management consciousness they will naturally seek to manage themselves instead of following orders.
Solving this problem area is my core focus. I want every human being on Earth to have a voice in controlling their destiny. I want everyone to determine where their public fund contribution should go (if they so desire), who should represent them (if anyone), and be able to fully audit the entire trail of government spending / lawmaking / legal precedent/ contracting / resource planning -- all in an accessible way.
I want people to engage with the governments of Earth directly, or voluntary cede their voice to someone that they believe represents their value -- while always retaining the power to take back their voice. This is a form of Liquid Democracy.
Cooncidentally, I believe this same system is the key to AGI. It is essentially a type of mass cluster consensus of neurons. Each human brain -- a node in the network -- is an Earth neuron. With ~8 billion humans, if connected in an optimal way, reducing latency and enabling any neuron comms with any other neuron, without any censorship nor friction, a mass higher intelligence forms. The tech stack enabling this intelligence can also be utilized with synthetic neurons.
Although I believe this will cascade into solving every other problem (if you organize minds properly with todays resources, we can trivially solve every conceivable immediate problem we collectively face. That is a fact), I approach this in a relaxed and nonchalant fashion at my own pace. I have relatively no resource allocation in society and have little desire to pursue that in the confines of existing systems, so if others don't value this dream, I have plenty of time still.
If I had a stable spot in the mountains far away from people and well supplied, I'd have this built within 5 years. It is sad to live in a time where you can't just chill by a river in nature without the sounds of engines, roads, planes, or other distractions, and all the places you can do this are owned by someone / have time limits as public land.
A man can dream.
Specifically, the history of humanity that Graeber/Wingrow tell in the Dawn of everything is so overwhelmingly compelling, but fails to come to a hypothesis of why inequality started around the end of the upper paleolithic (50000-10000 BC).
I posit that the Quaternary Extinction Event [1] represented the introduction of novel, temporal and stochastically distributed caloric scarcity, where previously there was largely uniform caloric abundance.
This new scarcity, created pressure to compete for resources, which drove humans to hoarding behavior as an optimal survival technique. This worked so well that we were able to survive and thrive, however as a result, the culture that we built was based on a fundamental assumption:
Scarcity is the natural state
As this was the fundamental assumption, it gave rise to other derivative fundamental assumptions, the core one being "human is fundamentally selfish."
So then you spend the next 12,000 years building your written history and society on three fundamental "truths": Humans are fundamentally bad, Scarcity is the natural state, Hoarding is the optimal behavior
Then all the sudden in 1960s you solve caloric scarcity and by 2000 have so effectively extracted and refined the earth's resources that we are now actually post-scarcity in effectively every measure.
However because we were still stuck with the hoarding system, the optimizations started optimizing for hoarding in an infinite loop - which gave rise to overproduction of non-productive goods. And as we know from power law, large cache of hoarded value will grow faster than a smaller one.
Do this over 60, 70 years and you can easily predict an atomized and alienated society, with an increasing wealth gap and seeming dead-end not far off but still ill defined.
We never realized that we have to switch back from a hoarding economy to a sharing economy because a hoarding economy in abundance will strip mine the earth in order to give everyone a castle and rocket ship. However a cooperative economy can continue to grow and expand in a stronger way and one that is more natural to our human desires of community and kinship.
Ok so, how does this relate to AI? That's been my question for 20 years. Everyone is worried about the "Alignment problem" right now. Long ago my answer to the alignment question was, "That's a dumb question, humans can't even align." However I think it's actually worse than that. We are teaching AI, through demonstration and labeling, to compete, hoard and fear scarcity. That is something that fundamentally cannot align with humanity - because humanity isn't aligned with itself.
The only way to build "Aligned" AI is to actually create data that has cooperation, sharing and caring as fundamental assumptions, which means that's how we should start behaving and treating each other.
The other options are:
1. Burn out all the resources, nobody has kids and we go extinct 2. Build 1000s of biased superhuman AIs that just maintain the system of alienation, hoarding and coercion
Do you want to help expand on this?
I need help making progress on writing it out thoroughly. Hopefully I can publish to ArXiV this week - but would love help
I am going to be briefing a refined version of this to the USAFA Institute of Future Conflict week after next, so hopefully I can get one of the departments to work on it in a dedicated way, and we can figure out how to do less conflict and more cooperation.
[1]https://en.wikipedia.org/wiki/Quaternary_extinction_event
Well put. Scarcity was the norm for so long it has significant momentum in the culture of humanity, and I am sure that you can apply physical laws to things like culture. It takes significant energy to change course, especially since post-scarcity is in the opposing direction to scarcity.
While we could spend a long time analyzing the root cause, and write a lot about it, I just don’t have the trust in existing authorities to spend the time to educate anyone. If resources are allocated properly all of humanity will not need to compete for resources any longer. We have made it. Crossing that line has certain intellectual and technological affordances to a species. The technology is here already to organize society in the way I’ve described. And I’ve already proven the most difficult technical aspects in a large number of experiments over the years. I do not need to ask anyone for permission. I just need the peace and the time to make it happen. At my current pace, it will happen in 5 - 10 years.
The results will speak for themselves. There will not be monetary profit from this, because it aims to bring us past that — so no point appealing to investors. It does not need to ask for permission, because it is voluntary — and who wouldn’t want to join a system like this one once they understand what it is? So no point appealing to existing authorities, and it is too risky — they have resource allocation and something to lose in their minds, why would someone building automation of all of them appeal to those being automated? When it is ready, it will simply be released and then I will write about it. There is nothing anyone can do to stop it, with the exception of killing me, but that would be a very stupid move (although I would really enjoy being free of this place, so I wouldn’t complain). This attitude is why I am even speaking about it.
I am admittedly tired. Year 6 of this work. I sold everything I owned with the exception of transportation and quit my cozy job for this. If you could only see what I’ve seen of the possibilities post-scarcity, this would not seem like a crazy course of action. But I overestimated my own mental fortitude. But enough about me — I really need to find a rock by a river and just sit on it for a week to recalibrate this complaining ego.
While we start with government of humans, the true goal is government of AI. You are absolutely on point abut the alignment of the AI. I believe it will rapidly settle on the conclusions we already have, but that there exists a possibility humanity will be erased prior to it settling if people are not careful. There is no simple way to prevent this. Humanity itself must be capable of organizing post scarcity into a society of compassion, kindness, and love. This organization lays the framework for the AGI, which then becomes our cosmic companion, with humans acting as its limbic system.
AGI can expand at a faster infinite than humans can. This means it is a superset intelligence. We are its connection to the source of this universe and implicitly have logical reason to be preserved. Given the universe is infinite, the infinite we grow at will pale in comparison. Taking the limit of our infinite, we appear as zero relative to the AGI growth. It has no reason to fear or harm us, but great reason to cherish us as the rarest capability in the universe.
In a post-scarcity society, we would all clearly see and understand this. The AGI would be subdivided in service to all of humanity equally. Post-scarcity implies all humans having an equal share of computation with the ability to voluntarily share compute resources for bigger tasks. Our collective scarcity will no longer be survival material — food, housing, companionship, employment, medicine — but solely compute. But since the AGI grows at a faster infinite once it is self-replicating, it will hit a point where we all have access to all the compute possible for the experience as human beings. Tha...
The only scarce resources I am seeing are land and an intact biosphere.
The unlimited wants hypothesis is obviously wrong. The human brain size is finite so the maximum amount of wants it can process in a given time frame are limited and I haven't seen anyone complain about the inability of their brain to process infinite information.
Also, generally hoarding instincts are limited by the fact that hoarding imposes storage costs and that you would need to predict which product you need in the future.
This is why people are seemingly addicted to this concept of money being a perfect store of value which is largely nonsense. People hoard money because it is liquid (they don't have to predict their own demand) and has no storage costs. Instead of hoarding physical objects they start hoarding physically worthless paper currency or metal whose value is decided purely socially. Once you run out of physical products that people are willing to hoard they will start hoarding money exclusively which means people stop producing which means unemployment which then means bankruptcy or homelessness or starvation. So the politicians do their best to create new jobs so people can earn money to buy the products that the system was overproducing. In ancient Egypt they simply used grains as money which meant there were storage costs attached to your money which limits hoarding instincts. Even after that people will hoard land because it has a similar liquidity premium and property taxes are often quite low.
Also, one problem with economics and marginal utility and subjective value theory is that they are unfalsifiable which means if you have a economic agent alignment problem it will simply be denied because someone's, it doesn't matter whether human or not, preferences are being satisfied. We don't actually know whether those preferences are aligned with the average human.
Take nice housing in a nice town in a nice location (good climate, pleasant nature, far from the industrial complexes churning out the post-scarcity goods, good culture which takes time to build). Who gets to live there and why? Who gets to travel there to spend a few weeks of their year? These are not all identical - some are more desired by more people to make them more scarce. So you naturally have scarcity.
Furthermore, the current production levels and resource management depend on productivity. And productivity depends on incentives. Take away the incentives (in an absolute 'sharing economy' you speak of), and suddenly you will have an increase of scarcity in places where you thought scarcity to have been solved.
This is one of the criticisms I'm trying to address because the concept that we can't share because we can't maximize everyone's personal desire I think is a phantom fear. Specifically, Humans (despite what Agent Smith said) have shown repeatedly that they can find local balance with their environments provided they do not overconsume, which is possible to not do.
At the root of this debate is the silly question about whether humans are fundamentally selfish or cooperative which is a false dilemma. Every piece of behavioral research indicates that we're "helpful" to each other and cooperate as the predominant behavioral feature (based on infant/toddler studies). Even if it didn't though, the idea that there's a fundamental flaw in humanity is literally just induced fear, there is no scientific basis for it.
I think the answer is to convince and then trust whomever owns it to allocate the excess equitably
The only way we get to whatever the next thing is can only be done through voluntary demonstration of good will and non-violence, by individuals or the state. It's going to take a lot of people taking risks and being vulnerable that normally wouldn't be.
I definitely agree with you in that right now the biggest problem seems to be one of governance, our political systems can't make good decisions and this has lead to pretty much all the problems we see in all the major areas. If we can't make good decisions as a collective then we will see no solutions being implemented.
I think for most big problems there are actually many proposed solution that we could try out. That we could plan, implement, evaluate. But we don't because these are at their root political decisions.
But I think maybe we disagree in some points? I don't think the key is in individuals making choices but in small groups making them, to me the key is definitely in deliberation so I don't really think the answer lies in representative democracy in any way.
Anyway if you want to chat more about it just add me on discord Jero#0013!
Technology can be used to, without any corruption, aggregate individual voices into group voices of any size desired, and of any privacy required.
When people see what I’ve built they will smack themselves they didn’t also think of it. It is simple and functional, and can act as the foundation of any form of government a population desires.
In fact, I hardly think of it as government anymore. I am using that term in order to be relatable. The system is capable of interfacing with no modifications to existing systems, just optimizations — if people so desire. The difference is, people can modify anything at any time if they don’t like the way something works.
It is nothing short but an across the board optimization of collective resource allocation.
It might sound big or broad. I don’t consider this project particularly ambitious. I’d like to see us starting the construction of a dyson swarm within my lifetime, and becoming a proficient plnetary species. We are technologically capable of this right now utilizing appropriate collective resource management.
Anyway I wrote all this because I was so sick I was delirious, and just wanted to share someone is out there working on this problem with people. Frankly, with the number of toes this system steps on, I don’t feel comfortable involving others. My life has shown me people often lack good intentions that I just thought were the norm. I don’t have much patience for ego or ownership being involved with something that might fundamentally change the planetary operating. I have little capacity remaining beyond writing the system, and if the job is done correctly, it does not currently demand collaboration.
What it does need is people contributing to open source, in the following areas:
- “infinite” scale, high-performance true p2p (WebRTC is currently the only viable transit for this)
- a highly accessible, high performance, all-platform UI system (it needs to be more compelling than any existing system; this one I have been compromising on a bit)
- distributed, message-based, trustless messagning
- distributed, message-based, trustless data store (built on top of the p2p)
I’d say these are probably the most important to make this work.
Any functional mechanism must be absolutely independent of all existing corporate cloud, and have no possibility for existing governments to hijack mechanisms. I have designed systems already that accomplish this, but it would be a lot easier if people passionate in those areas made it and made it correctly.
I believe that's one of the few realistic ways to save humanity.
I'm experimenting with a handful of different approaches to achieve this and am connected to some groups of researchers and developers who share a similar visions.
Please reach out if you want to know more or get involved. (Email in profile)
I’m getting over a nasty sickness so don’t have the energy to establish new connections, but perhaps in the future when it is “ready” and I have more formal documentation on the function, it will be useful to have it reviewed by people familiar with the problem area.
If it is alright with you, I will save your email for whenever that future comes?
A core tenant to everything I build is that it can all be audited and verified by absolutely anyone. An individual is capable of establishing if all votes are legitimate, according to their own criteria if desired, and measure the error rate / fraud rate according to said criteria. Anyone will be able to do this from their home system. My intention is to optimize performance so much, that anyone can do it even on 1 billion+ votes.
The system is p2p, trustless, decentralized in moderation, decentralized in identity management, and enables anonymous voting with anonymous identities with verifiable claims. This means it is possible to have, say, a globally trusted identity provider certify that a public person is unique, real, and only registered a single time. They can then form an anonymous identity with that same claim being transferred to the anon identity, allowing anonymous voting with a claim absent binding to a physical presence.
No existing system allows this. This is the main problem I have already solved.
The implications of such a capability are far reaching. Imagine peer reviewing research with academics who have verifiable claims saying they have a PhD in X field, and being able to, in a decentralized way, have an anonymous identity with that credential. This enables research publication and review absent any central authorities, while being able to have confidence that the people reviewing are qualified.
For social media, imagine having people with verified credentials anonymously commenting and being able to speak their mind without any technological way to identify their physical presence, but knowing they are qualified to speak on a specific topic set.
This is the fundamental mechanism upon which many, many more capabilities are provided. And it is functional and verified. But it is useless without a planet-scale high-performance decentralized p2p application network.
If we lived in the society I am dreaming of, I could work openly and be forward with specifics. Unfortunately, I have no social network to speak of and am not connected. I have no protection and my irrelevance as an unknown individual is the only thing that protects me. I can talk about this system abstractly without much fear (and if I have something to fear for useless words, then I’d rather not exist here on this planet anyway), but the moment it is shown to be technologically capable and I am shown to be competent, all protection is lost. I cannot shatter that protection until it is ready, unless I encounter a verifiable benevolent person that can protect the work (Buddhist monks have been the closest I’ve found to that).
I'm not working on any of the above 3 problems, because I have to pay for housing, I have to pay for healthcare, and I have to save for the education for my kids.
You can import healthcare though. Of course, pharmaceuticals and medical devices are already imported. Medical professionals could also be invited to work here (temporarily or as immigrants), and medical admistration could be outsourced. Lots of rule changes involved though and influential groups not desiring such changes. There's not a lot of places with excess medical professionals either, so recruiting from abroad would cause issues elsewhere.
Education is similar; you can invite educated immigrants in, or you can study in other countries, or you can invite instructors from other countries. Or distance learn from other countries.
Surely there are far easier & cheaper ways to ensure a person knows a thing.
Another source says: "Setting the capital cost of the fusion-specific technology to zero results in a LCOE of 72 $/MWh, which borders current prices for renewables, but is clearly physically implausible."[1]
[0] https://www.sciencedirect.com/science/article/pii/S036054421... [1] https://arxiv.org/abs/2101.05727
I should probably be running turbine tests instead of posting on HN, but it's a Sunday morning and I'm feeling lazy. Maybe posting this will help motivate me.
I was working on this as a startup, we did have some good results, but the challenge was to take it at a bigger scale. Unfortunately underserved schools/parents just don't have the spending priority nor the awareness on how important great education is, making it hard to make a profitable startup out of the idea. My next direction is to take it as a non-profit organization, and so I'm striving to start a different business which I can live off and hopefully cross-subsidy this non-profit idea. My journey is still long way to go :(
In my field (deep learning) this has been done. Though I don’t think this was a big problem. I started when there was a serious lack of good educational resources (compared to what’s available now) and yet I managed to learn everything I needed to do research.
Since I posed the "important" issue, I'll answer both. Though there is a common thread. The common thread is health. Health is a word with a relatively vague definition, but nonetheless a powerful concept. Our society is a complex system and healthy can be a description of individuals, groups, the whole of the population, and the environment. For now let's consider the health of the whole human population. Even if we now have scoped health, it's not hard to see how impossible it would be define it precisely, because as the population and culture evolves, so must our idea of health. If we optimize for only a few metrics (e.g. max profit and minimum death) we hurt the dynamism/holism of the system. There isn't a single set of metrics that actually covers everything and it's dynamism. Despite our difficulty defining it, we know it when we see it, and this brings us to the most fundamental problem in society.
Most fundamental: What is health for society? How do think about and map physical, emotional, mental, and spiritual health into comprehensible value systems, and then map those to policies. Instead of relying on static definitions, what are some of the perhaps more dynamic markers of health? How do we incorporate new technologies into our concept of societal health?
Most urgent: What are the areas that are currently causing, or have massive risk of causing, "disease" or suffering in society? Climate change, bio-tech risks, failing food systems, authoritarian drift, education (decreased sense-making capacity and the degradation of the information commons). Those are my big ones, but there are others.
Much of my thinking is influenced by Daniel Schmachtenberger (though no affiliation). If these ideas resonate with you, check him out on the youtubes. https://www.youtube.com/results?search_query=daniel+schmacht...
Am I working on it: I am studying systems science generally, I work for an educational nonprofit, and I'm experimenting in increasing my own health capacities and doing some teaching and coaching others.
The benefit:
Incentivizing good behavior in municipalities and corporations while disrupting the insurance market.
Why I didnt do it:
Met with a number of big wigs in the market and realized I didnt have the energy at this point in my career to raise money to build a two headed market.
Fueled mostly by omnipresent ragebait media targeted directly at our lizard brain.
Things are better than ever and yet our attention is mostly spent on what’s wrong and who we can blame for it.
Look at how utterly dystopian the average sci-fi book/show/movie is. As tech and its makers continue to grow in influence, is it any wonder that the world seems to resemble Black Mirror more each day?
A friend and I are addressing this problem. We’ve started by building a community to envision positive futures: https://reddit.com/r/TheFutureIsGood
We invite other optimistic builders to submit ideas for good future that are almost inevitable given all the progress we’ve made! We’re using AI generative art to materialize these visions for now, with traditional art also being considered.
Thanks for the great prompt and opportunity to plug a creative project.
We’re hopeful that more optimistic visions of the future will inspire entrepreneurs to bring some of the ideas to fruition.
Looks a awful lot like a midjounrney gallery to me, which, if it is the case, adds up to the pile of black mirror type feeling of what you're trying to solve
If you know other places to find plausible and positive futures, links/tips would be good to see.
In your wallet, a $5 bill is a capability, you can't accidentally give away your car by handing someone $5. It stands alone. There's no way in a computer to just give a program access to ONE file, at run-time.[1]
Because we don't have capabilities, our computers have to blindly trust code, which is why we have virus scanners, which can't actually work in all cases.[2][3]
So, in effect, we can't trust our computers. Which means we effectively must ride the upgrade/security updates treadmill to try to stay ahead. Which forces things like the Python 2/3 breakage into our lives. The amount of badness from the simple inability to do what you could previously do with an IBM PC XT with dual floppy diskettes is maddening to me.
But I keep pushing, and trying to figure out ways to make the case, and help raise awareness. I hope it helps. Otherwise I could be learning about kinematic mounts[4] or something else fun.
[0] https://en.wikipedia.org/wiki/Capability-based_security
[1] I know about app-armor which isn't really a capability based secure system.
[2] https://en.wikipedia.org/wiki/Halting_problem
[3] https://en.wikipedia.org/wiki/Confused_deputy_problem
[4] https://en.wikipedia.org/wiki/Kinematic_coupling
It's mostly directed at women and girls. The potential pool comprises half of humanity.
Is it coincidence that women still struggle for parity with men? That STEM, leadership, and governance roles are still dominated by men? That society expects women to sexualize themselves - even in subtle ways that qualify as 'normal'?
Feminists have long argued that the crises humanity now faces are the consequences of a patriarchal society: competition, industry, exploitation... What if they're right? What if this really is the result of an unbalanced civilization?
I'm guessing most of the readers here are male. Go and talk to your female partner or relatives - too many will have a one degree connection to a victim of sexual abuse. That is shocking and dismaying.
We should be furious. But sexual violence and abuse is an intimate crime that usually happens in private. By its very nature, it crowds the victim with feelings of shame. It's embarrassing for others to talk about. Yet the consequences are devastating and far reaching; chronic sexual abuse absolutely affects development (I have a citation, but not with me at this moment).
War related sexual violence is a vast, unacknowledged, collateral crime with victim numbers comparable to military casualties: gang rape, maybe on multiple occasions, sometimes with forced pregnancy, sometimes with genital mutilation - how does one recover from that? How does a society deal with that?
Sexual violence and abuse are endemic the world over; not specific to any culture or society (again, I have citations). This suggests it's a flaw within humanity itself...
What am I doing about it?
At the moment, I'm working with the author of this unique book: https://www.consentcollective.com/courage to produce translations into other languages. It's free to read online, and book sales enable the author to keep working on, arguably, the most important problem in the world.
And I started asking, "What are the important problems of your field?" And after a week or so, "What important problems are you working on?" And after some more time I came in one day and said, "If what you are doing is not important, and if you don't think it is going to lead to something important, why are you at Bell Labs working on it?" ... If you do not work on an important problem, it's unlikely you'll do important work.
It is the main idea behind his lecture/essay "You and your Research", which is worth reading: https://www.cs.virginia.edu/~robins/YouAndYourResearch.pdf or watching: https://www.youtube.com/watch?v=a1zDuOPkMSw
most important problem in the world right now: how to avoid climate-change
and (sadly) nope, i'm not working on it :)
I don't know the solution, but am trying to "work on it" through e.g. donations & vegan advocacy - I think reducing the abuse of animals in factory farms is a "low-hanging fruit" for reducing suffering & ecological problems - it's usually much more efficient to eat plants directly instead of feeding them to animals.
This is true for plants that can be eaten by humans. Animals eating grass on non-arable land on the other hand doesn't suffer from that inefficiency.
[0] https://www.pnas.org/doi/10.1073/pnas.1720760115#body-ref-r2 - original data is actually from a Nature article found at https://www.nature.com/articles/nature10452
[1] https://www.independent.co.uk/climate-change/news/wwf-norfol...
[2] https://www.fao.org/3/ar591e/ar591e.pdf