Yes, it's a great technical achievement, but I just worry for the future. We don't have good social safety nets, and we aren't close to UBI. It's difficult for me to see that happen unless something drastic changes.
I'm also afraid of one company just having so much power. How does anyone compete?
Realize that it's a choice to respond to things this way. This feeling comes from a certain set of assumptions and learned responses.
Remember that people are bad at predicting the future. Look at the historical track record of people predicting the implications of technological advancement. You'll find that almost nobody gets it right. Granted, that sometimes means that things are worse than we expect, but there are also many cases where things turn out better than we expect. If you're prone to focusing on potential negatives, maybe you can consciously balance that out by forcing yourself to imagine potential positives as well.
Try to focus on things you personally have control over. Why worry about something that you can't change? Focus on problems that you can contribute to solving.
It's easy to say "don't worry" if you haven't been affected by events like this. I feel it's stronger for society to say "I don't know what will happen, but we'll work through it together."
We are building technology, to suggest no agency is helpful in avoiding any feeling of responsibility or guilt — perhaps rendering your comment within the realms of waxing philosophical.
Who better to worry about this than the people of hacker news?
From a pure mental health standpoint, sure, it’s solid advice but I think it’s narrowed the context of the broader concern too much.
An alternative to learned helplessness of “nothing you can do” is to encourage technologists to do the opposite.
Instead of forgetting about it, trying to put it out of your mind, fight for the future you want. Join others in that effort. That’s the reason society has hope — not the people shrugging as people fall by the wayside.
Depression mediation by agency feels more positive, but I don’t have a lot of experience tbh. Just a view that we, technologists, shouldn’t abdicate responsibility nor encourage others to do so.
That culture, imo, is why a large section of tech workers, consumers and commentators see the industry in a bad light. They’re not wrong.
EDIT: to add, “what problems can I personally solve” also individualises society’s ability to shape itself for the better. “What problems can I personally get involved in solving”, “what communities are trying to solve problems I care about” is perhaps the message I’d advocate for.
But reality is as reality is and nobody is owed a desk job. These are very exciting times with what type of society could be built with this tech, human inefficiencies are responsible for a lot of suffering that we can might be ab;e to stamp out soon.
As an individual, I think the experience would be dreadful
Not in my thirties and almost nothing I worried about has come true. I mean, tomorrow we might get wiped out by a runaway technological singularity, but I could've spent the last 30 years of my life worrying a lot less too..
I went to cognitive behavior therapy, for me it was like someone opened up my mind and showed it to me on a screen, it was a mirror into my head. It was amazing how it felt like I could rewire my thought patterns over the course of a few months.
The main takeaway from it all however, was the mantra: _thoughts are not facts_.
If you can realize that your thoughts are not objective truths, you will be much better off in almost every aspect of your life, because after living this mantra for many years, putting it to the test constantly, I know it's solid.
Later on I read a lot of Buddhist philosophy which matched incredibly well with the therapy because a lot of Buddhist thinking and meditation practice is quite similar in it's approach. This sort of reinforced the validity of the CBT because I realized wise people have known about seeing things in an objective light for millennia, which was validating for me and helped me continue on the introspective path.
Basically, we're all hallucinating in one way or another, almost all of the time, and that is ok, just be aware of that. When we're worried about the future, we're worried about something which doesn't yet exist, which is actually crazy.
Of course it doesn't mean we should just ignore long term problems, no one advocates for that. But we shouldn't assume we know the outcome in advance because that often causes stress.
Warning: I think that for most westerners, it's "safer" to get into something like CBT, Buddhism comes with some IMO very confronting ideas for a lot of people where as CBT is much more user friendly for westerners.
Industrial scale technology might ruin us though, so you might have some point, mostly I'm referring to climate change which is for sure the greatest existential threat imaginable right now. However it seems technology might bail us out here too, nuclear and renewables.
Why do people always say this / think that saying this is helpful? Try saying to someone with ADHD, "realize that you are choosing not to get your chores done today. You're choosing not to get out of bed on time. You're choosing to show behavior that your peers describe as 'lazy'. This will keep happening as long as you let it!"
So what if you have the ability to choose whether you are depressed or not? Not everyone got the same choice. Not everyone still has that choice.
I don't really expect another solution, but this always kind of bothers me when I see people saying everything is a choice.
With neurodivergence and mental disorders, what you see as "choice" can end up not being a choice at all.
It's an easy thing for people to say when they don't really want to help others.
For example, I’m getting text messages all day long from random politicians asking for money. If you told people 50 years ago that one day we’d be carrying devices where we could be pinged with unwanted solicitations all day and night, they might have imagined an asphyxiating nightmare. But in reality, it’s mainly a nuisance.
The point is that your brain makes all kinds of emotional predictions about the future, but they aren’t really very useful and if you’re experiencing depression or anxiety, I can guarantee they are biased predictions.
The tens of thousands of people working in entertainment building other people's visions can now be their own writers, actors, and directors. And they'll find their own fans.
Studios will go away. Disney will no longer control Star Wars, because your kids will make it instead. In fact, the very notion of IP is about to drive to zero.
And OpenAI won't own this. They won't even let you do "off book" things, and that's a no-go for art. Open source is going to own this space.
There are other companies with results just as mature. They just didn't time a press release to go head to head with Gemini.
Only some people can make Star Wars (the pinnacle of independent filmmaking if you read Lucas's biography). It has nothing to do with the tools.
IP in the arts is how artists get paid.
I can assure you that no one in the creative industry feels liberated by these tools. Do you realise that just because you are good at lighting, you don't want to be an actor and make a movie? No, you like to be good at lightning, work with others who are good at what you do, and create a great work of art together.
AI imagery only knows what exists. It's tough to make it do innovative technical effects and great new lightning. "oh my god, stock video sites are dead" Yes, exactly; stock, by definition, is commoditised.
Edit: Putin on AI, 2017 https://youtu.be/aJELcvjREgk?t=29
"Whoever takes the lead in this area will be the ruler of the world."
"тот кто станет лидером в этой сфере будет властелином мира"
It's terrible news for the people being replaced. Their training and decades of experience is their competitive advantage and livelihood. When that experience becomes irrelevant because anyone can create similar quality work at the push of a button, they're suddenly left with nothing of value in a world flooded with competition.
Like billionaires didn't already scare a lot of people.
Dall-E was crazy and then suddenly people were doing the same thing on consumer hardware with an open model within a year.
Filmmakers being able to bring their vision to life using generative models is going to create such a huge expansion of the market.
What people don't realize is that long term these advances are a death knell for mega-corps, not for individuals.
Why do I need to kiss Weinstein's ass to get my movie made if I can do it with a shoestring budget and AI and have the same assistance to create marketing materials, etc. I need a lot less money to break even and can focus on niche markets aligned with my artistic vision instead of mass appeal to cover costs plus the middlemen involved in distribution and production.
I’m sure some people would be more ok to work a shitty job with the hope that they might make it as an artist.
Now that it’s becoming more and more obvious they will not be an artist that’s better than AI, what do they have left to hope for ?
I made a twitter thread[1] with weird metal cybertrucks using Midjourney a couple days ago. I personally enjoyed the process and do not have the talent nor the time to do that without generative AI. There are people who do have that talent, but honestly I doubt anyone else would've put in the time.
I think you might have it a little backwards. For most people, the fun part is "making a movie", not "watching hundreds and hundreds of hours of footage picking between 10 different shots". That's the drudgery, and that's the part generative AI can eliminate.
Most “good” art isn’t just what you see, it’s also the story behind it. Why was it made? What is the story of the artist? What does it make you feel?
AI might allow more people to tell some of those stories they may have lacked the raw skills to tell before. And for those who have the skills, they can make exactly what they envision, without being limited by some of the randomness in the AI. I think there will always be a place for that, and at the top of the market, that’s what people want.
Yesterday I asked a local llm to write a python script to have a several multimodal llms rank 50,000 images generated by a stable diffusion model. I then used those images to train a new checkpoint for the model and can now repeat the process ad infinitum.
In the olden days of 2020 I would have had to hire 5000 people each working for a day to do the same.
Are we though? People still do plenty of things out of interest or hobby, despite it being fully automatable?
e.g. blacksmithing or making certain homemade things?
While these are non digital things, why can't we apply the same thing here?
Some people still hand write assembly out of the novelety and interest of it. Despite there being better tools or arguably better ways of writing code.
But I agree, there's going to be a "war" around this and if could go wrong, if we are not careful.
There is no moat. This will all be commonplace for everyone soon, including with a rich open source community.
OpenAI won't let you do nudity or pop culture, but you can bet your uncle that models better than "Sora" will be doing this in just a few months.
> We are giving the enjoyable parts of life to a computer. And we are left with the drudgery.
No. This means that the tens of thousands of people working in entertainment building other people's visions can now be their own writers, actors, and directors.
This is a collapse of the Hollywood studio system and the beginnings of a Cambrian explosion of individual creators.
An AI capable of this would likely also be able to e.g. flip burgers and bring us to Fully Automated Luxury Communism.
The only original creativity will be in creating new formats and new kinds of experiences - which will mostly mean inventing new kinds of AI.
Everything made in an existing format will either be worthless or near as.
Same applies to software dev. Far more quickly than most people expect, it will also apply to AI dev.
And to hit scale first might be enough of a moat?
these advances are a death knell for mega-corps, not for individuals.
The torrents will be filled with convincing full-length films which are maybe not as good as the originals, but still very watchable.
We will have infinite chances to get better films than Indiana Jones and the Kingdom of the Crystal Skull.
*: except the ones starved for compute resources of course
I think we'll soon see a suburban mom in middle America with a part-time penchant for storytelling make a blockbuster video game mostly by herself.
Anybody can publish anything with the web. Have publishing houses disappeared? Have scientific journals disappeared?
The insanity is doing the same thing over and over and thinking "this time it will work".
In a world where attention is scare I sadly think big corps and power brokers will still play a large role. Maybe not though.
Of what market? Certainly not film production. I have my doubts about whether it will expand the market for films, in the economic sense. The lower the cost of producing and distributing a film, the lower the monetary value people will place on it.
Look how most music artists are no longer able to survive on royalties, and a few massive streaming companies have an astounding profitable oligopoly of consumers' music interest. Yes, many pre-streaming publishers were exploitative or unethical, but I'm not convinced that it was to a greater degree than the current market leaders. Consider also that the streaming revolution steamrolled many, perhaps most, indie record labels that supported niche genres; some live on but are no longer able to sustain physical output and a reduced to being digital marketing companies.
Now, people will continue to tell stories and entertain others, so technology like this will be good for people with an artistic vision who can't easily access publishers for whatever reason. It will certainly allow people to pursue bold artistic visions that would not otherwise be economically feasible - exotic locations, spectacular special effects, technically complex perspective moves. Those are good things; I worked in the film industry for a long time and have several unproduced scripts that I'd like to apply this technology to, so I'm not rejecting it.
However, more content doesn't necessarily translate into more economic activity; I think it very likely that visual media will be further devalued as a result. People who have spent years or a lifetime developing genuine craft will be told to abandon it in favor of giving suggestions to a computer system, and those who don't will be laughed at or suspected of fakery, because fakery is so widespread these days (Relevant recent example: https://news.ycombinator.com/item?id=39379073). The easier it becomes to make something, the less value the market will assign to it; rational from the abstracted perspective of pure price theory, disastrous in real life.
Increasingly, we seem to be tilting towards a Huxley-esque dystopia of stunning and infinite-feeling virtual worlds to which we can escape on demand, and an increasingly shitty real world marked by the brutal economic logic of total resource and information exploitation. Already a stock rejoinder to complaints about the state of things is that humanity is on paper richer than ever before, to the point that bums have smartphones and anyone can afford an xbox. I have a homeless neighbor who's living in his car, spending his dying years watching YouTube on his phone to fall asleep because he's lonely. Technically this is an expansion of the market, but I don't think it's a good outcome.
Medium term is ... less good.
AI turns skilled labor into cheap ones, supposedly, right? It's a massive enabler for mega-corps. Not a death knell.
Nothing about the future looks particularly good, other than that medicine is improving. But what’s the point of being alive in such a sanitised, ‘perfect’, instant-dopamine-hits-on-demand kind of world anyway?
Just say to hell with it and bury yourself in an interesting textbook. Learn something that inspires you. It doesn’t matter if ‘AI’ can (or soon will be able to) do it a billion times better than you.
And be kind to those around you.
1) My wife trained as a typesetter on a photo typesetting machine. That was already replacing typesetters working with lead, and the people sorting the used lead, and working with inks etc. They still needed a past-up artist and more. Eventually the GUI based computer arrived, with PageMaker, Quark, Indesign etc. These days she is super productive with a massive online icon library available, full printing and distribution capabilities. Able to do a job that could have involved half a dozen or more people previously.
Are those people unemployed now? Not really (we are talking 1.5 generation now, so not the same people). Unemployment levels are low, and the workforce is significantly larger, with men and women working. The working (outside of the home) part of the population has gone up significantly over a few generations, despite all the new enabling productive tech.
What I see is a lot of visually higher quality work being delivered, but often with the same core content. So productivity has increased, but you get a glossy new shiny report in a PDF, instead of a photocopy of a typewritten page. (Yes, I do simplify. But I think you get the gist.)
2) I started as a system admin, the a systems analyst, worked through project manager, etc. until I was leading startups. In the space where I work now, circular economy and food production, there is so much work to do, that any AI support we can get is welcome. But as the work is innovative, new and not done before, most often the AI tools aren’t that useful, yet. That may change, but with a society that needs to replace a significant part of the infrastructure and processes to achieve a long term sustainable society, I actually don’t worry the AI tools will take my job or any of my colleagues job away. There is plenty to do. I have enough new things in front of me that I could probably keep a whole big venture fund occupied for a long time.
"Now, you can typeset everything in your office" - early Macintosh ad.
needs to replace a significant part of the infrastructure and processes to achieve a long term sustainable society
One can adapt to needing to learn new technology, but one cannot adapt to an algorithm out performing you
The world is way bigger than technology and the Internet. It hasn’t really gone anywhere
Fresh air and sunlight are important though.
on the flip side, we'll just generate VR wilderness in the near future and nobody will care what's real or not
It's probably the same feeling farmers had in the beginning of the 20th century when they started seeing industrialized farming technologies (tractors, etc). Sure, farming tech eliminated tons of farming jobs, but they have been replaced by other types of jobs in the cities.
It's the same thing with AI. Some will lose their jobs, but only to find different types of jobs that AI can't do.
> jobs that AI can't do.
Sure, by definition, you’ve described the set of jobs that won’t be replaced by AI. But naming a few would be a lot more useful of a comment. It’s not impossible to imagine that that set might shrink to being pretty much empty within the next ten years.
Not remotely comparable. Farming is a backbreaking job, many were happy to see it going away. This is taking over the creative functions. Turns out what Humanity is best at, is menial labor?
But when one is 30+ years old, or even 40+ years old, it's hard to completely switch careers, especially when you're also dealing with the fact that it's not because you were bad at your job. Rather, a machine was made to replace you and you simply can't compete with a machine.
It's evolution, of course, but it is a stressful process.
I've started reading again, because reddit/instagram/etc. has become kind of boring for me? Like, I still go on them to get an instant dopamine hit from time to time, but like you said burying yourself in a textbook just feels so much more rewarding.
I’m sure such groups already exist, but maybe not specifically with this goal in mind.
Learning for its own sake really is the answer to lasting happiness… for some of us, anyway.
> Nothing about the future looks particularly good, other than that medicine is improving.
How do you reconcile your thoughts with what the CEOs of these AI companies keep telling us? I.e. "the present is the most amazing time to be alive", and "the future will be unimaginably better". I'm paraphrasing, but it's the gist of what Sam Altman recently said at the World Government Summit[1].
Are these people visionaries of some idealistic future that these technologies will bring us, or are they blinded by their own greed and power and driving humanity towards a future they can control? Something else?
FWIW I share your thoughts and feelings, but at the same time have a pinch of cautious optimism that things might indeed be better overall. Sure, bad actors that use technology for malicious purposes will continue to exist, but there is potential for this technology to open new advancements in all areas of science, which could improve all our lives in ways we can't imagine yet.
I guess I'm more excited about the possibilities and seeing how all this unfolds than pessimistic, although that is still a strong feeling.
Three ways:
* It's the job of CEOs to advocate the benefits of what they're doing.
* Those things might be true, for them.
* Those things might be true, from a global perspective, even if there are some people who are worse off. White-collar workers might just be those people worse off.
It gives me some hope, which fades rapidly as soon as I remember the vested interests such people have.
I'm just trying to not let it get in the way of appreciating the world. I'm planning to travel to mainland Europe sometime next year (gap year). SpaceX has reignited spaceflight, and there's so much cool stuff going on in that space. Science marches on, with a steady stream of interesting discoveries.
And programming is great - for now. It feels slightly strange spending a week writing a project that may be finished with a single prompt in a few year's time, but it's enjoyable.
Maybe I'm overreacting? I've grown up in a pretty calm period, with the west in a clearly dominant position. Maybe this is, paradoxically, a return to normality?
― Vladimir Ilyich Lenin
You are not alone : https://youtu.be/h3-va0umXTY?t=383
PS: youtube.com at 6:23, "Leonardo DiCaprio,,Julia Butters in Once Upon a Time in... Hollywood --break"
Just another tool.
Until full automated agent that is able to carry out a task from start to finish without human intervention, there is something for us to do I guess.
I doubt whether this is true. Lots of hype, but no tangible improvement to show for chronic conditions for common people.
- fast acting analog insulins that are metabolized in 2-3 hours instead of 6-7
- insulin pumps that automatically dose exactly the right proportion of insulin
- continuous glucose monitoring system that lets you see your BG update in real time (before, it was finger sticks 4-5 times a day; before that, urine test strips where you pee on a stick to get a 6 hours delayed reading (!))
- automated dosing algorithms that can automatically correct BG to bring it into range
In aggregate, these amount to what is closer than not to a functional cure for type 1 diabetes. 100 years ago, this was a fatal condition.
The systems that deliver the medical care might be, however (and indeed observably locally are, in many cases).
As for social safety nets: if this affects people as heavily as you think (on an unprecedented, never before seen level), the US will almost certainly put _something_ into place and add some heavy taxes on something like this. If tens of millions of Americans are removed from the work force and can't find other work because of this, they'll form a really strong voting block.
Also consider that things are never perfect. We've had wars around the world for a notable amount of time. Even the US has been in places we shouldn't be for a serious chunk of the last century, but things have worked out. We have a ton of news and access now so we're just more aware of these things.
Hopefully that perspective helps a bit. HN and social media can have "doomer" tones quote a bit. Hopefully some perspective can help show that this may not be as large a change as we think.
Or maybe I'm an idiot, as some child comments may point out shortly.
By definition, we don't have data for events we haven't seen before. So instead I reason as well as I can:
Consider the set of all jobs a human being could do. Consider the set of all jobs an AI system could perform as well as a human being but more cheaply. Is the AI set growing, and if so, how quickly?
Prior technology is generally narrow and dumb: I cannot tell my cotton gin to go plant cotton for me, nor can I ask it to fix itself when it breaks. Therefore I take on a strategic role in using and managing my cotton gin. The promise of AI systems is that they can be general and intelligent. If they can run themselves, then why do I need a job telling them what to do?
"Computer" used to be a profession, where people would sit and do multiplication tables and arithmetic all day [1]. Then computing machines came along and put all those people out of work, but it also created entire new categories of jobs. We got software engineers, computer engineers, administrators, tons of sub-categories for all of those, and probably dozens more categories than I can think of.
I think that there's a very high likelihood with the current jobs that humans do better than computers, most will be replaced by cheaper AI labor. However, I don't see why we should assume that set of things that humans do better than computers is static.
Most of my programming job is tightly coupled with the business processes and logistics of the company I work for, AI will not replace me there.
Also I'm not convinced this is sustainable, I'm thinking this will be like GCI where the first iron man film looked phenomenal but where huge demand + the drive to make it profitable will drive down the quality to just above barely acceptable levels like the CGI in current marvel blockbusters.
"I still haven't seen any text written by AI that doesn't contradict itself a few paragraphs later"
"I still haven't seen any picture made by AI that doesn't look like an abstract nightmare"
"I still haven't seen any picture made by AI that includes hands with the right number of fingers"
"I still haven't seen any videos made by AI that aren't janky and uncanny"
"I still haven't seen any videos made by AI that move me"
The transition from the first to the last took less than five years.
Wondering what the runway on this statement is.
When I use Unity I write ten lines of code and the tool generates probably 50k. Ever looked into the folder of a modern frontend project after typing one command into a terminal? I've been 99% dependent on code generation for ages.
If AI dystopia is coming, at least it's not here quite yet, so I'll try to enjoy my life today.
Every large animation studio has continually been looking for ways to decrease the number of artists required to produce a film, since the beginning of the field.
AI has no spark, no drive, no ambition, no initiative, no theory of mind, and it's not clear to me that it will ever have these things. Right now, it's just a hammer that can build 100 houses a second, but who needs 100 slightly wonky houses?
Also, 100 slightly wonky houses will sell like hot cakes if each one costs less than 1/100th of a not-slightly-wonky house. People will buy 100 of them instead of 1 and just live in a different one every day/hour so they always notice the novel parts instead of the wonky parts. We've had mass manufacturing for centuries and they always prevail when the trade-offs are acceptable.
So, it is completely alike what a lot of humans are like, at least at their jobs?
Just add a cost function.
I can see all of my plans for world domination coming together right in front of my eyes. A few years ago I was absolutely certain I’d die without achieving my dream of becoming God Emperor of a united Planet Earth.
Exactly. People just aren't seeing this. You don't even have to limit the fake memories to real people. Don't have a girlfriend? Generate videos and photographs of you and your dream girl traveling the world together, sharing intimate moments, starting a family. The possibilities are so exciting. I think the people who hate this idea are people who already have it all. They're not like me and you.
We have seen this in small scale of social media that ones self esteem.
We will see a new set of problems that would be much deeper. Videos and image that make you believe false reality, reliance of GPT will generate false knowledge.
False reality problems have started popping up everywhere. It is going to be much deeper. I think we are in for a really crazy trip
Advances like this are necessary steps to get us there.
People will be displaced in the short term, as they have for every other large scale advance... Cars, assembly line and so on. Better to focus on progress and helping those disaffected the most along the way
AI is going to be massively deflationary. How useful is UBI when the cost of goods and services approaches zero due to automation?
With that said, I can imagine the federal reserve will then helicopter in money to everyone in order to reach its 2% inflation target, which kinda sounds like UBI.
Right! I keep saying, that at least we have to kickoff the process. Not even the legislative process, but convincing the public that we'll need it eventually (alternatively a whole different system worldwide, but that will be even harder). Will take a long time anyway.
Many people (including myself) have bought into the narrative that history will repeat here and things will be better eventually, but not how much has to break first, and it's used as a hammer by OpenAI and probably every innovator who disrupted.
They advertised "Safety" but no "Economic Impact" analysis because the latter is less scary and requires difficult predictive work, the former is just narrow legalese defined by 80-year-old congressman they have to abide by to "release" v1.0. There is at-least a Congressional Budget office(CBO) where the 80-year-olds work, flawed as it maybe...
My fear is the alternative reality that these tools could provide. Given the power and output of the tooling, I could see a future where the "normal" of a society is strategically changed.
For example, many younger generations aren't getting a license at 16. This is for a variety of reasons: you connect with friends online, malls cost money, less walkable spaces, less third places.
If I'm a company that makes money based off of subscription services to my tools, wouldn't it be in my best interest to influence each coming generation?
Making friends and interacting with people is hard, but with our tooling you can find or create the exact friend you want and need.
We can remember now that life is beautiful - but what's to stop from making people think that the life made by AI is most beautiful?
And yeah, I've heard this argument before with video games, escapism, etc. I'm talking more about how easy it is to escape now, and how easy it'd be to spread the idea that escapism is better than what is around you.
For the most part, the idea of change is rarely inherently bad (even though, IMO, it's natural to inherently resist it) -- and humans adapt quickly to the parts that have negative impacts.
Humans are one of, if not the most, resilient race on the planet. Younger generations not getting licenses, sticking to themselves more, escaping in different ways, etc are all "different" than what we're used to, but to that younger generation it's just a new normal for them.
One day they'll be posting on HN2, wondering whether the crazy technological or societal changes about to come out will mean the downfall of their children (or children's children), and the answer will still be the same: no, but what's "normal" for humankind will continue to change.
In Europe there's no need. Got a licence over two decades ago have never needed to drive. Shops in walking distance, public transport anywhere in the country, convenient deliveries, walkable and cyclable cities.
Meanwhile other places have no freedom from cars, locked into expensive car financing, unable to access basic amenities without a car, and motorists have normalised killing millions of people a year.
More climate change, war, microplastics in our body and now extreme joblessness ?
If I woke up and I saw a headline that said OpenAI has developed and AI which told us how to sequester huge amounts of cO2 then I’d be excited and agree.
Exactly. I'm sick of people advocating changes as "progress" until we get some fundamental baseline sampling of humanity's well-being. When "are you depressed?" "do you contemplate suicide?" "are you exhausted?" go up for 10 years around the globe, then people will look like lunatics saying this is "progress" and maybe we'll have a better conversation about where progress actually is.
We had a half-assed lockdown for a few months where most people just kind of stayed indoors and saw noticeable environmental improvements world-wide. An unaligned AGI can easily conclude the best way to fix these problems is to un-exist all humans.
Yeah, progress has gotten us nowhere /s
There is some usefulness to those feelings - this announcement will probably have an impact on your life soon enough. But you cant let every button push and distant threat pull you down can you.
Also remember, life has its own ways: as far as you know, it could also be the beginning of the best days of your life.
We need to push back generally against our top heavy economy. More local and regional focus would help.
Btw, a year or two from now you'll be able to run a more powerful open model locally. So, not they aren't having some outsized amount of power
Yes some people will be scammed as they always have been, such as the recent Hong Kong financial deepfake. But no, millions of people will not keep falling for this. Just like the classic 419 advanced free fraud, it will hit a very small percentage of people.
IMHO humanity will be fine, decades from now kids will be asking what it was like to live before "AI" like how we might ask an old person what it was like to live before television or electricity.
I'm not optimistic.
2. I don't think what they have can be protected all that well. Others will catch up.
- paulg
As a side note I shudder to think how many nightmare fuel cursed videos the researchers must have had to work through to get this result. Gotta applaud them for that I guess.
> No one weaves now, and that's fine.
Did horses find new jobs when we moved to steam power? Leave aside the odd horse show and fairground ride. By the numbers, what do you think happened?
They found alternate employment as pack horses in WW1. The problem was solved after that.
I don't think anything will fix or change this, definitely not UBI, the situation is a fundamental part of the human condition. I share your dread and fear that I will not be able to compete, even if my life improves by all other measures.
The one thing that would dramatically change my calculus is medical advances that significantly push back death and aging.
I just dread, *shrug*. You don't have to be depressed or doomed, it all comes from premature predictions.
That said, we are surely at the phase similar to that one right before the internet, if not electronics. I genuinely don't understand people who write AI off as yet another Bitcoin or "just an enhanced chatbot". They'll have to catch up on an insanely complex area (even ignoring rocket science behind all that) which will do without them, and right now is the opportunity to jump the ship early.
It's only my nobody's opinion, but I can't see the way in which _that_ could fail. I find it incredibly stupid to think so and to just live your life as if nothing happens. If you're not the ruling class or a landlord, f...ing learn.
In the US, let's keep that in mind.
Industrialisation and computers/automation took away massive amount of jobs while globally improving people lives, this may possibly (maybe not) do the same.
If in the future, anybody can write the book, create a photograph or a motion picture or an music album with just few words describing what they have in mind, this will be a tremendous productivity improvement and will unleash an overflow of human creativity.
I like to compare it to what Jobs said about computer, they are the "bicycle of the mind" [1].
[1] https://medium.learningbyshipping.com/bicycle-121262546097
One reason is, readiness of tech does not mean it’s being applied.
Another is just like one OpenAI came out of no where, others will too. It’s normal to be focused on a few things to lose sight of that.
Gemini realistically does some impressive things.
What can we do?
Building the tech is important but applying it well for actual adoption is still wide open for the average persons use.
It does seem to mean that what we think might take 5y probably will take 1y in 2024, if not less like 2023. So think 10x, and 10x again as the real goal.
People will not loose their jobs, because you still need someone to input prompts for 10 hours straight in order to get a piece of video you want.
Natural language is not perfectly precise to get exactly what you want from a model like this, and the results remain kinda random. Instead of making video in a traditional tool you will be spinning AI roulette until it generates your desired result. And even then you will probably want to edit it.
Tools advance, but tools remain tools.
Open-source will catch up in <6 months. Note that Meta will ship llama3 anytime. So will Mistral.
I am a PM, and switched to becoming a builder. Enjoy learning, keep building. What people take time to realize is that building things is a habit. As you combine that with ongoing learning, you will enjoy the process and eventually build something to earn a living.
Unfortunately I have nothing to showcase yet.
For most of human history, people didn’t have constant access to art work or videos. Things were fine. Maybe instead of watching manufactured shit on social media, go see a live theatre production. Seek genuineness.
You can live a great life without ever seeing a picture or video.
Whether people are employed or not is a policy decision of the central bank and not related to how good AI is.
This stuff is going to change media and reality so much. Best to get involved in local groups.
Name one, and see if that holds up in 5 years...
Being personable, mediating conflict, leadership in general. Also cooking, baking and masonry.
I suppose it's going to be something like "You're asking for more purpose in your life? Sorry, I can't let you do that, Dave."
Also I think there will be more junk out there with how easy it will be to mass produce garbage.
https://en.wikipedia.org/wiki/CARES_Act
EDIT: For the downvoters, yes components of CARES was in fact inspired by UBI:
https://www.cnbc.com/2020/03/13/andrew-yang-aoc-free-ubi-cas...
https://www.businessinsider.com/coronavirus-aoc-demands-univ...
UBI has three basic properties that the CARES act fulfills none of
1. Covers cost of living for some basic standard (debatable, but should include food, water, and shelter at minimum)
2. Is available to everyone without onerous requirements or means-testing (IE is "universal")
3. Carries a reasonable expectation of continuity such that people can plan around continuing to have it
The CARES act was an emergency measure that absolutely zero people expected or intended to be permanent, it was laden with all the means-testing and bureaucratic hurdles that unemployment generally carries, and it very clearly did not provide adequate support for quite a lot of people
It's meaningless to call something a "test" when it carries none of the properties that proponents of a policy claim would make it desirable. The only perspective from which the comparison even makes sense is from that of someone who's not considered it seriously and come up with a strawman to argue against it (IE something like "UBI is the government gives people some money")
It also seems worth mentioning that I really don't buy the highly political claim that some people seem to view as self-evident: that people remained unemployed longer because they got extended unemployment benefits, rather than as a result of the massive economic shock that prompted that decision in the first place
To me, it seems guaranteed that will be drastic changes. There will be many attempts at new ways of organizing society with successes and failures along the way. Not out of altruism or desire to share but out of self-interest of those who collect the power afforded by AI and automation.
Just wait.
One company having that much power is a different matter, and I address it by looking at how we can distribute GPT training through decentralized and open platforms.
It won't be me or you. Whoever it is, they will not share any of the economic upsides of AI with the public unless they are legally forced -- zero, zip, zilch, nada. Even then, they will keep the lion's share for themselves, and they will use their surplus to shape society to their advantage.
So yes, many millions of us have a big problem to worry about, especially considering how much struggling there already is now.
But even if it stays in private hands, one company monopolizing a technology and keeping it expensive/out-of-reach is generally not how technological innovation works. There is generally intense competition between providers, with each aggressively cutting prices to capture market share.
Pure theology
We're already witnessing this with the creation of textual and graphical content through ChatGPT. It's now possible to generate various types of text content and a wide range of graphics at the cost of 10 cents for the dozen ChatGPT API calls. And the work is completed in a few minutes, as opposed to several hours. This represents a several orders of magnitude increase in per capita productivity for these specific tasks. As AI technology advances, the scope of applications benefiting from such productivity boosts is expected to widen, which means human civilization will experience a revolutionary increase in productivity, and with it, resource abundance.
I for one, would be overwhelmed. In the meantime I will be passionate and joyful about the things I like regardless of whether AI can do them a million times better. I have fun doing it.. while the AI is.. just AI.
You could be the best at using the tools, but I think there could be a point where there is no need to hire because the tools are just that good.
I am hoping our society and civilisation doesn't implode as everyone becomes unemployable as the economy and social contract collapses
Current AI wave is a corporate funded experiment desperate to find something compelling beyond controlled demos to economically recoup the deepening hole in their balance sheet. The novelty has begun to wear off, the innovation has started to stagnate and the money running out. The only money making innovation left to be seen is in creating more spam. Thats where I see this wave headed.
OpenAI has proven it's a shit company with rotten fundamentals playing with a shiny new toy. They will crash and burn spectacularly. As many before have done in various fields.
My reaction after using any AI tool from the last couple years to do anything meaningful ends with just a big facepalm.
Its a grift[1]
The forces of blind or cynical techno-optimism accelerating capitalism may feel insurmountable, but the future is not set in stone. Every day around the world, people in seemingly hopeless circumstances nevertheless devote their lives to fighting for what they believe in, and sometimes enough people do this over years or even decades that there’s a rupture in oppressive systems and humanity is forever changed for the better. We can only strive to live by our most deeply held values, within the circumstances we were placed, so that when we look back at the end of our lives we can take comfort in the fact that we did the best we could, and just maybe this will be enough to avert the inevitable.
“Glad did I live and gladly die, And I laid me down with a will … Here he lies where he longed to be; Home is the sailor, home from sea, And the hunter home from the hill.”
OSS so far has been an effective tool as a check and balance against big corp.
Touch grass, tend your garden, play with your kids, drink a beer, bake a pie, write a poem, take a walk, carve a sculpture, play a board game, mend a sweater, take a breath. Relax.
we will finally free ourselves from mediocre humans being the bottleneck for everything
If you want a safety net move to Scandinavia
So there are a lot of things to be depressed about before you get depressed about a little increase in misinformation and idiocy on the interwebs. I mean... things like polio and the measles are literally back to fuck with us because people are so fucking stupid they think vaccines are a bad thing.
It'll be fine.
If anything the optimist in me is hoping that all this "AI" generated content is going to make the internet so useless that our society (well the part that doesn't believe the earth is flat and that Bill Gates has mini clones in the vaccines) finally get away from it. In my region of Denmark our local police posts their immediate updates on twitter, which was fine when everyone could see them, not so great now that you need an account. I very rarely care about what they post, but around new years a fireworks container blew up near here, and I had to register (and then later delete) a twitter account to figure out if I had to worry about it or not. It'd be nice if the impending doom of fake content is going to move our institutions and politicians away from big tech SoMe platforms and it just might if they become useless.
These huge updates are interestingly timed though - same day?
Many people, rightfully, (over-)react to the American caricature of Christianity (mega churches, Kenneth Copeland, etc.) as the definition of what it is (that's arguably the deception hinted at in the Bible), but reading/trusting the raw word—what's referred to as "sola scriptura"—is remarkably helpful in navigating what's taking place.
…said every doomsday preacher since the Bible was written.
As does the I Ching
These are all just Rorschach tests, why choose one of the most corruptible and corrupted approaches?
> to the American caricature of Christianity
This cannot be overestimated.