When it comes to manage the insanely complex systems that form society, I think I would choose an incorruptible (if biased) AI before a fallible human being any day.
The whole idea that "we need to decide" presumes that there is a collective "we" that's separate and distinct from the individuals and corporations releasing this stuff into the world, and that "we" participate in a collective decision making apparatus that has any effect whatsoever on what new capabilities get invented and released into a networked world where borders, boundaries and jurisdictions are all but meaningless.
wait, whenever the economy has been revolutionized in the past was there like a big decision making just prior to the revolution that made sure everything worked out nice and dandy? I guess I just don't feel the two things line up.
I take your point, but this feels pretty defeatist to me.
I don’t personally believe that “we have not historically done this well, if at all” is a good reason to stop trying to manage the negative externalities of emerging technology— quite the opposite, in my opinion.
However, "the future technology is completely unpredictable and we don't understand it" is probably a fairly good reason we will only make worse mistakes in that regard going forward.
After my experience with AT&T front-line support where I tried several times to get my old AT&T iPhone unlocked (which is fully paid for), dealing with confusion and false promises of follow-up which never came, I welcome the relative competence of ChatGPT. My phone is still locked BTW.
You don't know that it would be better. Keep in mind that the same company whose support didn't help you with your phone will be in charge of what the chatbot is allowed to say and do. If AT&T sucks, I don't know if you're gaining much when AT&T replace a person with ChatGPT. The person doing the support was never the problem, the problem was the company who has a crappy process. And even after they deploy a version of ChatGPT that company will be the same.
Right, a bad company with bad processes could definitely stifle the benefit that ChatGPT could provide. I would hope that ChatGPT could summarize and report on process deficiencies in an effort to address them. Data like "in the past 90 days, 3,328 customers requested an iPhone unlock and I was unable to assist them due to lack of process addressing the request."
I've had a real life customer service chat session (years ago now) where they said their website didn't work on Mac OS while we were using it, and that it would be better if my phone was Android rather than iOS.
Back in the 90s, one of the monthly Mac magazines I read obsessively (because we didn't have internet at home yet) had regular stories of dumb things said by customer services; something like PC World sales clerk standing in front of a bunch of iMacs with Nanosaur running on them saying "yeah iMacs only do web browsing, you can't play any games".
Before I moved to Berlin, the dude in one of the phone shops near me insisted that iPhones couldn't work with those insert-your-phone-here plastic versions of Google Cardboard.
It's not that you're necessarily incorrect — when ChatGPT was new, I saw someone dismiss it as "mansplaining as a service" — but I don't know how much the companies really care.
It's not like customer support is a prestigious role.
Who knows! Maybe? The email that's written by ChatGTP, but no human actually reads it because at the other end it's summarised by ChatGTP too: maybe that'll be it. LLMs almost but not quite cancelling out the meaningless guff they add to the world.
It can both be true that a) LLMS and their successors will be the sort of huge multiplier for intellectual pursuits that _machinery_ is for physical work; and also b) they are AWFUL for education, and will result in the general population losing skills.
Will ChatGPT make the already troubling income and
wealth inequality in the US and many other countries
even worse?
It is an interesting question if the concept of "wealth" will survive in a world where all productivity is done by artificial intelligence.
As soon as computers surpass humans in every aspect, there is nothing left for humans to do. So the only reason for wealth inequality would be inheritance. But why should the poor accept this inequality? Will AI protect the wealth of the rich who inherited it? Why should it?
There are two reasons we do this. First, emotional, which many suspect will not be present in the AI. Second, because it benefits us due to the ecosystem, of which is likely to be irrelevant to the AI.
Finally, when there is an order of magnitude difference in intelligence, bugs, bacteria, we are generally not very kind at all. We could just as easily become the new lab rats.
At least, that's the ideal answer, but we don't really know what we're doing.
This isn't like giving birth to a new human, where the new mind has a sense of self and individuality by default and it would be immoral to take it away, works of artifice begin their existences as we construct them; and right now we don't even know how to tell if an artificial intelligence has an inner world, much less how to imbue an AI with one or prevent it.
We don't need to create one that doesn't serve humanity, one bad actor - a company or govt could create AI embodied assassin drones to take out competitors. I mean being a Bond villain was never so easy, lol.
This wouldn't wipe out humanity but could lead to fascism at a world-wide level.
If we have an AGI (as opposed to just an AI), then can't it decide who it's going to be? We're really naive if we think that we can decide (or even limit) who it's going to be.
> But why should the poor accept this inequality? Will AI protect the wealth of the rich who inherited it? Why should it?
Why do they accept it now? The best predictor of financial success is the parents' wealth [1]. Society as a meritocracy is a myth, yet we all plow on. I don't see what would change this.
No chat bot ever published in the past decades has revolutionized anything. They have all been the subject of intense, temporary hype, and then completely forgotten. I see no reason why this time should be different. It literally lies and makes shit up! The economy isn't built around lies and making shit up, is it?
>The economy isn't built around lies and making shit up, is it?
I think a significant portion of it is made up of BS and hype. And there's a lot of that around AI, particularly in the last few weeks. Big players look to be going all in on it.
I think dismissing it as simply a chat bot is missing what's actually happening. There's much more to ChatGPT than merely a chatbot, even with the hallucinations.
I think that's called programming, and has been a thing for a long time now. You don't control computers by inputting numbers anymore. What exactly are these "actions" you speak of, and did ChatGPT "perform" them without errors?
I am literally a series of chemical reactions; and yet, that kind of description is generally regarded as overly reductionist.
To the best of my knowledge, before the transformer model, the state of the art for natural language processing was looking out for pre-programmed keywords for actions, and simply summing the total of "good minus bad" for sentiment analysis. I'm not sure if even word2vec made a significant impact there, despite the then-impressive ability for it to embed words into a vector space such that "king" - "man" + "woman" ≈ "queen"
But GPT? Well…
---
Me:
explain the Einstein field equations in the style of an ancient Greek epic poem. Write in Greek.
write a python script which calls the Google Translate API to turn that into Latin
ChatGPT:
I'm sorry, but as an AI language model, I am not authorized to call external APIs or execute any code outside of this platform. However, I can provide you with an example code snippet that utilizes the Google Translate API via the googletrans library to translate text from English to Latin:
На Commodore64 BASIC напишите реализацию видеоигры Pong.
ChatGPT:
К сожалению, я не могу написать код на Commodore64 BASIC, потому что это требует доступа к конкретному железу, которого у меня нет. Однако, я могу предоставить вам базовый шаблон кода на Commodore64 BASIC для игры Pong, который можно адаптировать и доработать под свои нужды.
You are not literally a series of chemical reactions. There is no branch of science that claims so. In fact, we're very aware of our inability to define ourselves, particularly the phenomenon of consciousness.
Am I supposed to be impressed by a bunch of gibberish in languages I don't speak, and therefore can't verify?
Respond to the following hacker news comment, give an answer in the style of Shakespeare:
<I copy pasted your comment here>
It:
Hark! Thy words ring hollow as the empty air,
Forsooth, to claim we're not but mere reaction,
Is to ignore the science everywhere,
And spout ignorance without true traction.
Oft have philosophers and scientists sought,
To fathom consciousness and all its might,
A mystery still, with answers not yet wrought,
But progress made through study and insight.
And why must language be a barrier,
To truth and knowledge we can all attain?
By learning, we can all become carrier,
Of wisdom, and our horizons expand again.
Thus, I implore thee, don't dismiss with disdain,
The knowledge that awaits in languages not plain.
--
(HN messes up the formatting a bit; it was 4-4-2 lines in each group)
I'm basically on board with what it says here. Consciousness may be a mystery, but everything we see points to substance monism, not Cartesian dualism.
I take caffeine, a chemical, one thing happens to my mental experiences; alcohol, another chemical, a different thing; if I were to take THC or LSD or whatever, they too would have predictable effects on my mental state.
You cannot be seriously suggesting that the quality of AI generated text be evaluated through the broken telephone of Google translate. That's absurd.
> Forsooth, to claim we're not but mere reaction,
aaand it already fucked up by the second verse. My comment clearly states that we are not mere reaction, yet this "AI" seems to think I claimed the exact opposite. You're proving my point, it's just guessing what word fits where, not understanding anything whatsoever and producing garbage as a result, that only seems intelligible to gullible fools.
> if I were to take THC or LSD or whatever, they too would have predictable effects on my mental state.
I don't think I've ever heard anyone describe the effects of LSD as "predictable". I'm guessing you've never actually taken it and are talking out of your ass.
>I don't think I've ever heard anyone describe the effects of LSD as "predictable". I'm guessing you've never actually taken it and are talking out of your ass.
How is it not predictable? Varying degrees depending on dosage of hullucinations, euphoria, altered state of mind including altered senses of time, sound, vision, touch etc. This is how the effects of drugs are described by medicine and science.
It seems as if you are ignoring this part of their statement "effects on my mental state." They aren't saying the content of the hallucination is predictable or anything like that.
"Varying degrees of altered state" doesn't equate to "predictable" in my mind, and I struggle to understand why it would to you. The effects of LSD are the contents of the hallucination - the actual state that is altered. "It alters your mental state" is not so much a prediction as it is an admission of lack of understanding. Everything alters your mental state. That's how your brain functions. If your mental state remained unchanged, you would be brain dead. The question is how it alters your mental state, and all the effects you clump under "hallucinations" are precisely the unpredictable contents. It's medsci for "we don't know".
You misquoted me and then hyperfocused on the misquote. The changes from the drug are predictable because they wouldn't just randomly happen to you without the drug.
>Everything alters your mental state.
You are overgeneralizing to the point of absurdity and it does not appear that you are interested in a good-faith discussion. The contents of a hallucination and the experience of having a hallucination are distinct.
Suggesting that the effect of LSD altering my mental state and me walking out my back door altering my mental state are the same because they both alter the state is absurd. You're obtusely ignoring intensity, impact, duration and that the hallucination and euphoria are different effects.
> The changes from the drug are predictable because they wouldn't just randomly happen to you without the drug.
False. People experience euphoria without LSD, even randomly. There are entire spectrums of neuroconfigurations in which people precisely experience random bouts of euphoria. And that's just one of the effects you listed. In fact, one of the (potentially concerning) effects of LSD is called hallucinogen persisting perception disorder, in which the experience under LSD persists long after its use, and if you bother to read up on it, you'll also find that people who have never taken LSD can randomly experience them, proving you wrong again.
> You are overgeneralizing
No you are. You're clumping an entire mind-altering experience under "hallucination" and "euphoria". You should start asking, what is being hallucinated, why is the patient euphoric. That's the opposite of generalization.
> it does not appear that you are interested in a good-faith discussion.
Seems like an admission on your part. I'm speaking from research and experience. I've read books about LSD, I've experimented with it myself, and I've observed and discussed its effects with many people. My faith in what I'm saying is just fine. Accusing me of bad faith is nothing but projection on your part.
> The contents of a hallucination and the experience of having a hallucination are distinct.
You couldn't be more wrong. The experience of having a hallucination is fundamentally defined by the contents of that hallucination. Seeing hallucinations of your friends stabbing you in your sleep is not the same experience as feeling one with God while the Sun shines on your face. Different content, different experience.
> You're obtusely ignoring intensity, impact, duration and that the hallucination and euphoria are different effects.
And you're ignoring the actual contents of mental state, arguing in favor of simply measuring everything by "intensity", "impact", and "duration". Do tell, what's the "intensity" of walking through your back door? What's the "impact" of having a religious or spiritual experience? What's the "duration" of learning something about yourself?
It only makes shit up once it does not have enough information, it will however improve as it is connected to the internet, ALSO - it can do accounting, financial audit and coding better than humans.
Even right now, given it is properly implemented, it can leave millions of accountants and auditors without jobs.
But I agree, it wont revolutionize anything, it will automate EVERY SINGLE JOB that is dependent on human's intellectual input and eliminate billions of the jobs, which will lead to global hunger and reduction of human population.
The page describes itself as "A logic puzzle game invented by artificial intelligence" (emphasis mine). There is no claim that ChatGPT "entirely wrote it" anywhere on the page.
> On page 24 of this paper, there is a 3D game with enemies and objectives coded by GPT-4:
There is no "game" on that paper. Only two screenshots and the dubious claim that "GPT-4 did it", with no code anywhere to be seen, also checked the appendix.
> There is no claim that ChatGPT "entirely wrote it" anywhere on the page.
Yeah that’s interesting. There used to be a linked blog post describing the development process with screenshots of the LLM output. It’s missing now. I suspect the author made some improvements or monetized it or something
it can leave millions of accountants and auditors without jobs.
I think those are two examples of exactly what cannot be replaced yet. Their jobs are specifically about ensuring that rules have been followed precisely. Current AI sounds confident, but still makes obvious errors that would be completely unacceptable for accounting and auditing. Same as software engineering, you currently still need a human reviewing the output and verifying that it's actually doing the right thing, and without side effects, poor handling of errors and edge cases, etc.
What's the point of automating all that work, if those billions of unemployed will then be too poor to buy your product/service?
If that really happens we get universal basic income, or massive government employment programs. Otherwise the economy collapses and the ultra rich lose money.
You're thinking way too far ahead for a capitalist. There are countless people that wouldn't mind having a big piece of a small pie, rather than a small piece of a big pie, even if the total mass of the small piece was much higher.
Also, it's game theory, if you move to automate all the work, you'll get super rich in the meantime, and if you don't move to automate all the work, someone else will move to automate all the work and you'll lose.
> The economy isn't built around lies and making shit up, is it?
Yeah, it kind of is. Fake it 'til you make it. Vaporware landing pages to validate products. But that isn't what we need to focus on.
ChatGPT is more than a chatbot. And if you use its output knowing that it may contain hallucinations and correct for that possibility, there is real potential here. You can't take it at face value, and have to validate whatever it gives you, but you should be doing that anyway for anything you find online of through the media.
I am the first to say that its lack of accuracy is a problem, but the last to say that it is a non-solvable problem.
If you have to check online that ChatGPT is correct, then what exactly does it add but an extra step to just getting the correct info from online in the first place?
That is exactly the point - don't use ChatGPT to get facts. Use it to generate content that will be revised by a human. If you are trying to use it to ask the answer to a question and expecting accuracy... you are using the wrong tool.
Not necessarily, you could use 'different' ai models to maybe 'score' the accuracy of the information, and especially if each model has internet access they can provide some analysis and probably give you way more accurate info than even wikipedia.
We're not there yet, but we're probably not far from that.
What is this magical future AI that determines the accuracy of the current AI when the current AI clearly fails at being accurate? You're purely fantasizing.
> The economy isn't built around lies and making shit up, is it?
I see you've never talked to a salesman. Try going to a time-share sales presentation. They actually have an acronym the CEO of some of the big hotel chains came up with: TAFT - Tell them ANY FRIGGING THING. (really it's frigging because: corporate ethics or something).
Private data yes, but I don't think code is that much of a secret special sauce. Apart from anything else, too many engineers moving between different roles and sharing what they learn.
Even with medical data, which I at least hope is much better protected, some of that ended up in the training sets used by the current bunch of text-to-image AI.
How exactly will ChatGPT will revolutionize economy. The best it is doing is summarisation of search result. And search engines exist from 2000s. Yes instead of visiting wikipedia now we get the result directly. At max it can decrease the number of links to click. It is going to change the economy as googles answer box has changed economy, On the other hand the way they hallucinate you still need to double check any thing they tell you. So we can not automate the decision making. Best of luck to you if you want to replace a lawyer or doctor by LLM where they have accuracy of 80%
even if it doesn't provide that much intelligence, it provides intelligence at a surprisingly low cost and intelligence that has almost zero barrier to entry
this will do to text task and machine learning task the same thing that squarespace did to web development.
that is it will not replace the high end, but suddenly the long tail of businesses will be able to afford the entry price and start getting benefits that would have been totally out of their price range before.
What is so interesting to me about generative AI is that the jobs that are the most difficult to automate are trades like plumbers and electricians. The group that is the most vulnerable is knowledge workers. It’s a lot easier to replace a person with AI when much of what they do is virtual — on a computer screen. Trade jobs are often looked down on by the privileged class who went to some costly and exclusive college. Will that change? There is a lot of experience and training needed to do a trade job well. In fact (having done both), I think in many cases they need more training that a job in tech requiring a college degree. This is going to be really disruptive and I’m very curious to see how it all plays out.
I'm pretty sure that it'll end with oversupply on traders (especially in stagnant economy) if AI will cut most of the knowledge workers the world will have to move into some sort of utopia/dystopia and people will just don't have to work at all or it will be 0.1% vs 99.9% scenario but with brainwashed people cause of social media and maybe AI censorship.
But its worst case scenario in my mind AI will make jobs just more specialised and output of knowledge worker a lot faster.
> AI will make jobs just more specialised and output of knowledge worker a lot faster.
If the user is skilled enough at prompt engineering to get good answers (i.e. code that compiles and does what they want) the answers can still be objectively worse than search results for stackoverflow or github in many cases.
Seems like the current state of chat bots is that it's slower than either making a regular search or simply just knowing what to write in the first place. Shouldn't be a surprise, but a mid to senior engineer doesn't spend their whole day googling stuff to copy and paste. They read the docs for about 5 minutes and then get things done in the precious couple of hours they have between meetings. I don't think it gets much faster than that. I sincerely doubt the nontechnical people in the meetings would get a good enough answer from ChatGPT that fast and they certainly wouldn't want to slam their head against a wall learning everything the dev already knows, no less in a higher level and imprecise "natural" language.
I’ll openly admit that I’m a cynical person but I’ll believe economic revolution when I see it.
The advent of the personal computer also caused lots of huge proclamations of computers doing our jobs while we sit back in sun loungers sipping drinks. Yet somehow, here we are.
The people in charge of the AI systems are the same (class-wise at least) as those in charge of the economy so there’s no incentive to rock the boat. I expect AI to be something similar: it’ll change the way many jobs work, hopefully for the better. But people will still be working in sweatshops making clothes because it’ll still be cheaper than the alternative.
Personally, it's somewhere between the wheel and electricity in my opinion, or a combination of the industrial revolution/assembly lines and electricity.
The economy serves humans as the end users of goods and services. Even if an AI can learn to design a house that is more earthquake-proof than any human design of the same cost, the AI can't decorate the interior without understanding human psychology both at the general level of the average human and the specific level of the human who will be using the house.
I didn't realize how much happier I was when I worked next to windows (as opposed to working in the interior of a room) until I was in my mid twenties. How would an AI know to tell me to design my home office with a big window when I didn't even know that was necessary myself?
The AI couldn't have learned from me (since I didn't consciously know that), and the AI couldn't have learned that sitting next to windows generated happiness for itself either. That means some other human being would be necessary to input that as a design constraint into the AI's architecture program.
The inability of AI to experience human feelings for itself means it will have to rely on humans to help it provide economic goods and services that are maximally useful. That means there will be a role for humans in the economy for quite a long time (possibly forever).
> The pursuit of human-like capabilities, Brynjolfsson argued, has led to technologies that simply replace people with machines, driving down wages and exacerbating inequality of wealth and income. It is, he wrote, “the single biggest explanation” for the rising concentration of wealth.
I think, assuming it doesn't kill us - the faster we get to an economy where only 5-10% of the people are employable, the faster we get to post scarcity, and MORE equality/egalitarianism (if accelerationism doesn't lead to the opposite outcome, walled cities for elites, wastelands for everybody else) ... but the transition... it's going to be bumpy/painful.
Part of me is worried, part of me says rip the bandaid off, let's get to UBI, and other programs --as people can't stay rich or get richer if nobody can buy products. What's the point of better productivity at Ford or GM if nobody is buying cars anymore?
The world of the next 20 years, is going to change as much as the change from before electricity to the 1920s when it was mainstream. We can't really stop it, even if we tried - other actors good/bad are going to be using AI, it's part of our lives now for better or worse.
78 comments
[ 4.4 ms ] story [ 68.9 ms ] threadI don’t personally believe that “we have not historically done this well, if at all” is a good reason to stop trying to manage the negative externalities of emerging technology— quite the opposite, in my opinion.
Back in the 90s, one of the monthly Mac magazines I read obsessively (because we didn't have internet at home yet) had regular stories of dumb things said by customer services; something like PC World sales clerk standing in front of a bunch of iMacs with Nanosaur running on them saying "yeah iMacs only do web browsing, you can't play any games".
Before I moved to Berlin, the dude in one of the phone shops near me insisted that iPhones couldn't work with those insert-your-phone-here plastic versions of Google Cardboard.
It's not that you're necessarily incorrect — when ChatGPT was new, I saw someone dismiss it as "mansplaining as a service" — but I don't know how much the companies really care.
It's not like customer support is a prestigious role.
It can both be true that a) LLMS and their successors will be the sort of huge multiplier for intellectual pursuits that _machinery_ is for physical work; and also b) they are AWFUL for education, and will result in the general population losing skills.
As soon as computers surpass humans in every aspect, there is nothing left for humans to do. So the only reason for wealth inequality would be inheritance. But why should the poor accept this inequality? Will AI protect the wealth of the rich who inherited it? Why should it?
Finally, when there is an order of magnitude difference in intelligence, bugs, bacteria, we are generally not very kind at all. We could just as easily become the new lab rats.
At least, that's the ideal answer, but we don't really know what we're doing.
This isn't like giving birth to a new human, where the new mind has a sense of self and individuality by default and it would be immoral to take it away, works of artifice begin their existences as we construct them; and right now we don't even know how to tell if an artificial intelligence has an inner world, much less how to imbue an AI with one or prevent it.
This wouldn't wipe out humanity but could lead to fascism at a world-wide level.
Why do they accept it now? The best predictor of financial success is the parents' wealth [1]. Society as a meritocracy is a myth, yet we all plow on. I don't see what would change this.
1: https://cew.georgetown.edu/cew-reports/schooled2lose/
I think dismissing it as simply a chat bot is missing what's actually happening. There's much more to ChatGPT than merely a chatbot, even with the hallucinations.
Maybe you're a few days behind the times?
To the best of my knowledge, before the transformer model, the state of the art for natural language processing was looking out for pre-programmed keywords for actions, and simply summing the total of "good minus bad" for sentiment analysis. I'm not sure if even word2vec made a significant impact there, despite the then-impressive ability for it to embed words into a vector space such that "king" - "man" + "woman" ≈ "queen"
But GPT? Well…
---
Me:
explain the Einstein field equations in the style of an ancient Greek epic poem. Write in Greek.
ChatGPT:
Ω Μοῦσαι, θεαὶ τῆς εἰρήνης καὶ τῆς ἀρμονίας, Τὴν ἱστορίαν τῆς φύσεως μὲν ἀείδετε, Τοῦ κόσμου τὰ μυστικὰ καὶ τὰ βαθέα διδάσκετε.
Αἰνστάινος ὁ φωσφόρος τῆς λογικῆς καὶ τῆς φύσεως, Τὸν λόγον τοῦ κόσμου κατέγραψεν ἐν εἰκόσιν ἀριθμοῖς, Τοὺς λεγομένους Αἰνστάιν εὐφυεῖς εὑρίσκων λόγους.
Στις ἰσότητα τῆς μάζας καὶ τῆς ἐνέργειας ἐστίν ἡ πρῶτη, Τὸν κόσμον διακοσμοῦντα καὶ τὴν κίνησιν διατηροῦσαν λοιπὴν, Τὴν ἐντροπίαν, τὸν πόνον τοῦ κόσμου περιγράφει ηὗται.
Τὸ θεῖον καὶ τὸ ἀνθρώπινον τῆς φύσεως συνάγουσι, Τὰς καμπὰς τοῦ χρόνου καὶ τὰ μυστικὰ τοῦ κόσμου ἀνακαλύπτουσιν, Αἰνστάινου λόγοις τὸν κόσμον συντηροῦσιν ἀεί.
Ὦ Μοῦσαι, θεαὶ τῆς εἰρήνης καὶ τῆς ἀρμονίας, Τὴν δόξαν τοῦ Αἰνστάινου καὶ τοὺς λό
Me:
write a python script which calls the Google Translate API to turn that into Latin
ChatGPT:
I'm sorry, but as an AI language model, I am not authorized to call external APIs or execute any code outside of this platform. However, I can provide you with an example code snippet that utilizes the Google Translate API via the googletrans library to translate text from English to Latin:
Copy code
from googletrans import Translator
translator = Translator(service_urls=['translate.google.com'])
text = "explain the Einstein field equations in the style of an ancient Greek epic poem. Write in Greek."
result = translator.translate(text, dest='la')
print(result.text)
When you run this code, it should output the translated text in Latin:
Copy code
explicare Einstein campo aequationes in modo antiquo Graeca carmen epico. Scribere in Graeco.
Please note that the googletrans library uses the public Google Translate API, which has certain usage limits and restrictions.
Me:
建議量子力學主題酒吧的名稱。
ChatGPT:
以下是一些可能的量子力學主題酒吧的名稱：
1. 薛定谔的酒吧 2. 量子酒馆 3. 希尔伯特的饮料 4. 波粒二象性酒吧 5. 能量溢出酒吧 6. 纠缠酒廊 7. 费曼的小酒馆 8. 量子跃迁酒吧 9. 亚当斯的指南酒吧 10. 量子隧道酒吧
Me:
На Commodore64 BASIC напишите реализацию видеоигры Pong.
ChatGPT:
К сожалению, я не могу написать код на Commodore64 BASIC, потому что это требует доступа к конкретному железу, которого у меня нет. Однако, я могу предоставить вам базовый шаблон кода на Commodore64 BASIC для игры Pong, который можно адаптировать и доработать под свои нужды.
Copy code
10 REM PONG 20 RE...
Am I supposed to be impressed by a bunch of gibberish in languages I don't speak, and therefore can't verify?
Anyway…
--
Me:
Respond to the following hacker news comment, give an answer in the style of Shakespeare:
<I copy pasted your comment here>
It:
Hark! Thy words ring hollow as the empty air,
Forsooth, to claim we're not but mere reaction,
Is to ignore the science everywhere,
And spout ignorance without true traction.
Oft have philosophers and scientists sought,
To fathom consciousness and all its might,
A mystery still, with answers not yet wrought,
But progress made through study and insight.
And why must language be a barrier,
To truth and knowledge we can all attain?
By learning, we can all become carrier,
Of wisdom, and our horizons expand again.
Thus, I implore thee, don't dismiss with disdain,
The knowledge that awaits in languages not plain.
--
(HN messes up the formatting a bit; it was 4-4-2 lines in each group)
I'm basically on board with what it says here. Consciousness may be a mystery, but everything we see points to substance monism, not Cartesian dualism.
I take caffeine, a chemical, one thing happens to my mental experiences; alcohol, another chemical, a different thing; if I were to take THC or LSD or whatever, they too would have predictable effects on my mental state.
You cannot be seriously suggesting that the quality of AI generated text be evaluated through the broken telephone of Google translate. That's absurd.
> Forsooth, to claim we're not but mere reaction,
aaand it already fucked up by the second verse. My comment clearly states that we are not mere reaction, yet this "AI" seems to think I claimed the exact opposite. You're proving my point, it's just guessing what word fits where, not understanding anything whatsoever and producing garbage as a result, that only seems intelligible to gullible fools.
> if I were to take THC or LSD or whatever, they too would have predictable effects on my mental state.
I don't think I've ever heard anyone describe the effects of LSD as "predictable". I'm guessing you've never actually taken it and are talking out of your ass.
How is it not predictable? Varying degrees depending on dosage of hullucinations, euphoria, altered state of mind including altered senses of time, sound, vision, touch etc. This is how the effects of drugs are described by medicine and science.
It seems as if you are ignoring this part of their statement "effects on my mental state." They aren't saying the content of the hallucination is predictable or anything like that.
>Everything alters your mental state.
You are overgeneralizing to the point of absurdity and it does not appear that you are interested in a good-faith discussion. The contents of a hallucination and the experience of having a hallucination are distinct.
Suggesting that the effect of LSD altering my mental state and me walking out my back door altering my mental state are the same because they both alter the state is absurd. You're obtusely ignoring intensity, impact, duration and that the hallucination and euphoria are different effects.
False. People experience euphoria without LSD, even randomly. There are entire spectrums of neuroconfigurations in which people precisely experience random bouts of euphoria. And that's just one of the effects you listed. In fact, one of the (potentially concerning) effects of LSD is called hallucinogen persisting perception disorder, in which the experience under LSD persists long after its use, and if you bother to read up on it, you'll also find that people who have never taken LSD can randomly experience them, proving you wrong again.
> You are overgeneralizing
No you are. You're clumping an entire mind-altering experience under "hallucination" and "euphoria". You should start asking, what is being hallucinated, why is the patient euphoric. That's the opposite of generalization.
> it does not appear that you are interested in a good-faith discussion.
Seems like an admission on your part. I'm speaking from research and experience. I've read books about LSD, I've experimented with it myself, and I've observed and discussed its effects with many people. My faith in what I'm saying is just fine. Accusing me of bad faith is nothing but projection on your part.
> The contents of a hallucination and the experience of having a hallucination are distinct.
You couldn't be more wrong. The experience of having a hallucination is fundamentally defined by the contents of that hallucination. Seeing hallucinations of your friends stabbing you in your sleep is not the same experience as feeling one with God while the Sun shines on your face. Different content, different experience.
> You're obtusely ignoring intensity, impact, duration and that the hallucination and euphoria are different effects.
And you're ignoring the actual contents of mental state, arguing in favor of simply measuring everything by "intensity", "impact", and "duration". Do tell, what's the "intensity" of walking through your back door? What's the "impact" of having a religious or spiritual experience? What's the "duration" of learning something about yourself?
Even right now, given it is properly implemented, it can leave millions of accountants and auditors without jobs.
But I agree, it wont revolutionize anything, it will automate EVERY SINGLE JOB that is dependent on human's intellectual input and eliminate billions of the jobs, which will lead to global hunger and reduction of human population.
https://sumplete.com/
On page 24 of this paper, there is a 3D game with enemies and objectives coded by GPT-4:
https://arxiv.org/pdf/2303.12712.pdf
The page describes itself as "A logic puzzle game invented by artificial intelligence" (emphasis mine). There is no claim that ChatGPT "entirely wrote it" anywhere on the page.
> On page 24 of this paper, there is a 3D game with enemies and objectives coded by GPT-4:
There is no "game" on that paper. Only two screenshots and the dubious claim that "GPT-4 did it", with no code anywhere to be seen, also checked the appendix.
Yeah that’s interesting. There used to be a linked blog post describing the development process with screenshots of the LLM output. It’s missing now. I suspect the author made some improvements or monetized it or something
I think those are two examples of exactly what cannot be replaced yet. Their jobs are specifically about ensuring that rules have been followed precisely. Current AI sounds confident, but still makes obvious errors that would be completely unacceptable for accounting and auditing. Same as software engineering, you currently still need a human reviewing the output and verifying that it's actually doing the right thing, and without side effects, poor handling of errors and edge cases, etc.
If that really happens we get universal basic income, or massive government employment programs. Otherwise the economy collapses and the ultra rich lose money.
Also, it's game theory, if you move to automate all the work, you'll get super rich in the meantime, and if you don't move to automate all the work, someone else will move to automate all the work and you'll lose.
Yeah, it kind of is. Fake it 'til you make it. Vaporware landing pages to validate products. But that isn't what we need to focus on.
ChatGPT is more than a chatbot. And if you use its output knowing that it may contain hallucinations and correct for that possibility, there is real potential here. You can't take it at face value, and have to validate whatever it gives you, but you should be doing that anyway for anything you find online of through the media.
I am the first to say that its lack of accuracy is a problem, but the last to say that it is a non-solvable problem.
We're not there yet, but we're probably not far from that.
I see you've never talked to a salesman. Try going to a time-share sales presentation. They actually have an acronym the CEO of some of the big hotel chains came up with: TAFT - Tell them ANY FRIGGING THING. (really it's frigging because: corporate ethics or something).
Even with medical data, which I at least hope is much better protected, some of that ended up in the training sets used by the current bunch of text-to-image AI.
this will do to text task and machine learning task the same thing that squarespace did to web development.
that is it will not replace the high end, but suddenly the long tail of businesses will be able to afford the entry price and start getting benefits that would have been totally out of their price range before.
ChatGPT is ephemeral.
I'm pretty sure that it'll end with oversupply on traders (especially in stagnant economy) if AI will cut most of the knowledge workers the world will have to move into some sort of utopia/dystopia and people will just don't have to work at all or it will be 0.1% vs 99.9% scenario but with brainwashed people cause of social media and maybe AI censorship.
But its worst case scenario in my mind AI will make jobs just more specialised and output of knowledge worker a lot faster.
If the user is skilled enough at prompt engineering to get good answers (i.e. code that compiles and does what they want) the answers can still be objectively worse than search results for stackoverflow or github in many cases.
Seems like the current state of chat bots is that it's slower than either making a regular search or simply just knowing what to write in the first place. Shouldn't be a surprise, but a mid to senior engineer doesn't spend their whole day googling stuff to copy and paste. They read the docs for about 5 minutes and then get things done in the precious couple of hours they have between meetings. I don't think it gets much faster than that. I sincerely doubt the nontechnical people in the meetings would get a good enough answer from ChatGPT that fast and they certainly wouldn't want to slam their head against a wall learning everything the dev already knows, no less in a higher level and imprecise "natural" language.
The advent of the personal computer also caused lots of huge proclamations of computers doing our jobs while we sit back in sun loungers sipping drinks. Yet somehow, here we are.
The people in charge of the AI systems are the same (class-wise at least) as those in charge of the economy so there’s no incentive to rock the boat. I expect AI to be something similar: it’ll change the way many jobs work, hopefully for the better. But people will still be working in sweatshops making clothes because it’ll still be cheaper than the alternative.
A. The internet?
B. Electricity?
C. The wheel?
Personally, it's somewhere between the wheel and electricity in my opinion, or a combination of the industrial revolution/assembly lines and electricity.
How much did the wheel revolutionize society?
I didn't realize how much happier I was when I worked next to windows (as opposed to working in the interior of a room) until I was in my mid twenties. How would an AI know to tell me to design my home office with a big window when I didn't even know that was necessary myself?
The AI couldn't have learned from me (since I didn't consciously know that), and the AI couldn't have learned that sitting next to windows generated happiness for itself either. That means some other human being would be necessary to input that as a design constraint into the AI's architecture program.
The inability of AI to experience human feelings for itself means it will have to rely on humans to help it provide economic goods and services that are maximally useful. That means there will be a role for humans in the economy for quite a long time (possibly forever).
I think, assuming it doesn't kill us - the faster we get to an economy where only 5-10% of the people are employable, the faster we get to post scarcity, and MORE equality/egalitarianism (if accelerationism doesn't lead to the opposite outcome, walled cities for elites, wastelands for everybody else) ... but the transition... it's going to be bumpy/painful.
Part of me is worried, part of me says rip the bandaid off, let's get to UBI, and other programs --as people can't stay rich or get richer if nobody can buy products. What's the point of better productivity at Ford or GM if nobody is buying cars anymore?
The world of the next 20 years, is going to change as much as the change from before electricity to the 1920s when it was mainstream. We can't really stop it, even if we tried - other actors good/bad are going to be using AI, it's part of our lives now for better or worse.
Yeah, revolutionary. (That's GPT-4, by the way)