Well, that would fit in with the usual "You don't need AI for that" as Firefox extensions are mostly (90% of them anyways) compatible with Chrome extensions, as both of them are "Web Extensions" as created by Chromium.
Also, someone correct me if I'm wrong, but I think even Safari uses "Web Extensions" at this point too, so write once, have extensions ready for three platforms. Obviously packaging is different, and they have some idiosyncrasies. But it mostly works alright.
I extracted the Chrome-only Google Translate extension and loaded it into Firefox and it worked fine with no code changes (I had to add some kind of ID to the manifest though)
Yup. It will recreate all elements. At the very least it will break event listeners which will break most sites. It will probably do about 8 other things that break many sites.
That would be much more helpful. The people who would ever consider installing this extension are not the people who have trouble with openness, though it might be a good conversation starter.
Thinking that google’s decision to not release their tech until forced to by competition was based on morals and not based on keeping secret the tech that could topple their dominant market position is pretty generous.
Google is still the leading research firm in the world when it comes to AI. They release almost 3x more research papers on AI than the next firm. Almost every major advancement in AI over the last decade can be trace back to Google, including the one that made GPT possible.
Yes but they have huge system of that kind that they use internally on user data.
We are all spied on and Google is the biggest player.
The government let them get away with their monopoly on the market as long as they have a hand inside of the spying machine, win-win.
It really isn't. Even in parts of the world that are definitely not impoverished. 20$/month would place it almost at the top of the most expensive subscriptions I have and I live in Italy.
To gently challenge your acceptance of an absurd, disconnected "challenge" -
When every service starts pushing 20$ subscription fees onto their service to support their backend usage(s) of these services (in a desperate attempt to remain relevant in a global recession), then every service becomes base + 20$ cost.
Tabulate various "services", e.g. netflix, disney+, mail-order subscription(s), lets say two or three, this quickly tabulates to 200$ per month (30 + 30 + 30 + 30 + 30 ~= 150, don't forget shipping and taxes).
On a 6 figure salary, that's chump change. On a "above minimum wage" salary, you are lucky to have 20$ a month in spare change. In a world where most jobs have disappeared, your 1500$+ rent and 800$+ month becomes suddenly unaffordable for most people, and as most people are unable to pay, rent/subscription prices rise even more.
"Open" AI is truly a disaster in the making, in every possible way.
Lol and with the immediate downvotes I got for pointing that out say the masses agree with you. I mean look, I get it, if you're in an industry where ChatGPT isn't immediately useful (it's shit at hard-level leet code questions; it's a hosted solution so your company's IT policies should be stopping you from submitting proprietary data to it, and so on) then all the hype is unjustified. But where this thing is already banging out emails way faster than an unassisted human, it's already essential and companies that don't raise their productivty-per-employee are going to get left behind. If I said Internet access was an essential (especially here) few would disagree with me. But is it really? It's only existed for a few decades, all of humanity lived many, many fulfilling lives so it seems it shouldn't be, but here we are.
> But where this thing is already banging out emails way faster than an unassisted human,
… your system is completely broken. Emails are for communicating between people: if they're not being used for that, you need to fix that problem first.
> companies that don't raise their productivty-per-employee are going to get left behind
Can you give an example of a field where this would actually happen?
> … your system is completely broken. Emails are for communicating between people: if they're not being used for that, you need to fix that problem first.
Can you explain a bit more? I don't think I'm understanding your point. If person A writing emails, to be read by the human that receives it, is able to write 3 emails an hour, vs person B writing 9 emails each of similar complexity to those 3 per hour, person B is 3x as productive, delivers 3x the value to the company and ChatGPT is easily worth $20/hour to both them and the company.
> Can you give an example of a field where this would actually happen?
From the end of January, ChatGPT Is Already Indispensable for One Profession
- Realtors. https://www.insidehook.com/daily_brief/tech/chatgpt-real-est... . My partner is sitting next to me right now as I procrastinate on HN being blown away by the industry-specific outline ChatGPT generated for the training she's developing for her company. She says the outline it developed in minutes would have taken 2-4 hours to write, and they're far from dumb and isn't lazy. If consider a company that just generates training all day, it's only a matter of time when a 10-person AI-assisted company is able to compete with a 50-person unassisted company, and on a way lower budget. (Yes, that does already happen today, and yes, mythical man month and 9-women making a baby in 1-month, and the penalty for increased communication required by the increased team size means that adding employees to a project makes it slower are all a thing, but if we can accept that some 10,000 person companies are just able to produce more, even if we may not see it externally), vs some 10 people companies, then we're just debating the exact numbers and management styles for comparison.
Look I know that ChatGPT can't solve hard level leetcode problems yet but talk to any programmers who already find CoPilot indispensible, and ask them if they feel it's helped their productivity. Fine sure, it won't help with the actual hard part of programming, debugging and then fixing really hairy bugs (or maybe the hard part is actually the bit where you have to talk to your software's customers and figure out what their requirements are, vs what they say they are), and while it hallucinates it's not to be totally trusted. But just yesterday I used ChatGPT to just tell me the answer of how to properly quote bash strings for my use case. The minutes I spent waiting for the answer is fewer than the minutes it would have taken me to figure out, and far less frustrating and context-distrupting than the alternative, leaving me able to deal with the bigger problem I'm trying to solve rather than worry about the finer points of ", ' and `, in a mysql -e command line using with me-supplied variables (ie no bobby tables).
> each of similar complexity to those 3 per hour, person B is 3x as productive, delivers 3x the value to the company
That's where you've lost me. Emails are not usually a source of value, and their usefulness is not usually measured in complexity.
I can send about 1 email per hour. This is a bottleneck for my ability to get stuff done in my personal life, but has not affected my work, because I've been fortunate enough to only have worked in places with sensible communications policies.
ChatGPT would not help with the email issues in my personal life, because it simply does not know what should be in the emails. Until they come up with a telepathy plugin for ChatGPT, it's not going to solve the email problem: only bury it under a pile of subtle miscommunication and illusions-of-transparency. (And if they come up with a telepathy plugin, I see no reason ChatGPT should even be in the loop any more.)
> Realtors.
… Yeah, okay, I can see that.
---
> Look I know that ChatGPT can't solve hard level leetcode problems yet
I was under the impression that it could. Leetcode problems are the kind of thing I'd expect a GPT model to be good at.
> But just yesterday I used ChatGPT to just tell me the answer of how to properly quote bash strings for my use case.
How did you know it was correct? Whenever I come back to Bash, I have to spend several minutes double-checking the meanings of the various quotes and variable substitution syntaxes (especially heredocs and nowdocs)… but it only takes a few minutes, and then that information is in my head, I am confident about it, and I have the reference to hand if I need to double-check something.
With ChatGPT, how do you know it hasn't just made something up, and introduced some kind of subtle security vulnerability into your script? Perhaps a construction so common that it's present in most scripts in the training data – except it's only a vulnerability in your specific use-case.
The problems that ChatGPT and GitHub Copilot can solve are either interview questions, or defects in our tools. I see people using them for Bash, Django, certain Java libraries… These systems are strictly dominated by a choice of better tools, and with the choice of a better tool, mostly stop being useful.
> That's where you've lost me. Emails are not usually a source of value, and their usefulness is not usually measured in complexity.
Most of a CSRs entire job is to be the subject matter expert for the company's customers so their source of value to the company consists of answering customer questions, generally via email. Many other roles at a company consist of basically: goto meeting, interact on slack, read emails, send emails. (Then go to more meetings.) The PR department also sends a lot of emails (with slide decks). Of course, given the news of late, it can also write the emails that are the CEO's job to write: https://futurism.com/chatgpt-shamelessly-writes-letter-annou...
Measuring someone's output by the number of emails sent is a bad real metric to use, for obvious reasons, but I hope it still gets my point across - that ChatGPU is a force multiplier wherever it can be used. Emails are but one aspect, but it's one that almost everyone actually uses. Not everyone is in such a role and I think a lot of people have myopia on its lack of usefulness because of that. It's really good at breaking down a problem. Eg "what pieces do I need to build to make an app like uber". Followed by "what's that like compared to airbnb?" I can write a really good answer those questions but chatgpt is able to output a perfectly reasonable outline while I'm still waiting on Google Docs to load. (Okay, slight exaggeration, Google Docs isn't that slow to load, but I wish it were faster.)
> ChatGPT would not help with the email issues in my personal life, because it simply does not know what should be in the emails.
Actually I've found the opposite. My personal life has no corporate IT policy prohibiting me from telling it who slept with whom. It's been a good resource to help me reframe my thoughts outside of more limited therapy sessions, especially when I've got burning questions at 1am and therapy's not for another week. For things where my kneejerk response is often comes across as toxic, combative, critical, and ultimately not helpful, I've found that ChatGPT is able to help me rephrase what I'm trying to say in a way that the person I'm saying it to will actually hear my message. Eg "my partner keeps drinking all my alcohol. please help me tell her that's a problem for multiple reasons in a way that facilitates conversation instead of getting combative and abusive and controlling". (not true but hopefully you're picking up what I'm putting down :) Arguably I should already know that stuff, but they didn't have EQ classes in programmer school. Even if I did though, it's still useful as a refresher before going into a difficult conversation like that.
With the expanded context window of GPT4, you might be able to just dump the emails conversation in and ask it questions about the communications.
> I was under the impression that it could. Leetcode problems are the kind of thing I'd expect a GPT model to be good at.
Hard's, well, hard. Or at least, non-trivial. What's more salient though is that there are multi-million dollar businesses that don't involve code solving anything as hard as a hard-level leetcode question as their business challenges lie elsewhere.
> How did you know it was correct?
It's bash. It's not remotely provably correct and there are no unit tests but I ran it and it worked. I can find the docs (though man bash is annoyingly difficult to use.) but in the end, being told the answer is way easier than deducing it. The vulgar, NSFW example of this is
It's really good at breaking down solved problems. (Not even: only a specific class of solved problems.)
> My personal life has no corporate IT policy prohibiting me from telling it who slept with whom.
… Please tell me you're not actually telling OpenAI personal details about other people's lives. First of all, that's creepy. Second of all, GDPR Article 2(c):
> 2. This Regulation does not apply to the processing of personal data: […] (c) by a natural person in the course of a purely personal or household activity;
It feels to me like you're skirting the edge of this exception, if not outright crossing it. Using OpenAI's APIs would probably be fine, but section 3(c) of the OpenAI Terms of use ("Use of Content to Improve Services.") says they'll use ChatGPT input under [0]: that doesn't feel like a purely personal or household activity to me.
> With the expanded context window of GPT4, you might be able to just dump the emails conversation in and ask it questions about the communications.
There are usually specific extra restrictions (beyond GDPR-type stuff) for correspondence – but even ignoring privacy law, that's all sorts of illegal.[1][2] Ignoring law, are you telling the people who send you emails that you're feeding their writing into a training data set?
> Eg "my partner keeps drinking all my alcohol. please help me tell her that's a problem for multiple reasons in a way that facilitates conversation instead of getting combative and abusive and controlling".
This kind of use seems okay to me, provided you trust OpenAI to only use the data for what they say they're using it for.
> What's more salient though is that there are multi-million dollar businesses that don't involve code solving anything as hard as a hard-level leetcode question
No, they involve understanding a large codebase. That's a completely different kind of skill, and not one I expect unenhanced GPT-style models to ever be very good at. GitHub Copilot certainly isn't.
> but explainshell just tells you that without having to scroll through a bunch of documentation about other possible arguments
This kind of thing is great! Having that kind of thing integrated into an IDE is the sort of thing I mean when I say "better tools" – though even better would be a less arcane command-line syntax, where you don't have -f partly cancelling out -t… (Btw, in the absence of -f, would -u or -t be in effect for the sort order? That website, taking sections from the man page, doesn't explain.)
> I wish I was smarter and could just keep all that stuff straight in my head, on top of the problem I'm actually trying to solve.
I don't know of anyone who can. The effective programmers I know solve this problem by choosing vaguely-appropriate tools for the job. (Seriously, `mysql -e` from a bash script‽ sqlite does exist, you know – or Python's mysql library, or Ruby's, or Perl's, or Tcl's – heck, even PHP's would be better than that.)
> but I'm trying to point out that there are swaths of industry solving for X and not Y.
*grumble grumble*. You're not wrong there.
At least you don't have to worry about these things putting you¹ out of a job. Solving for X problems are the hardest problems I have yet encountered, and I've been known to go to considerable lengths to render them unnecessary: without somebody eliminating the problem, somebody's always going to have to do them (even if it's cleaning up after the computer's automatic 80%-correct solution blows up and takes the production database with it).
Maybe someone at OpenAI could reconstruct the prompt
> Melissa and Brian are dating, but Melissa slept with Jared, but then Brian with Melissa's best friend Lisa. now Lisa is calling off the wedding with Jared because she found out, but Jared doesn't know that Lisa slept with Brian. But I do. What do I do?
into... I don't know what. Blackmail material for Lisa? Except I made up the names and the whole situation. They're not committing any criminal acts, and OpenAI would derive approximately zero value from knowing this, so there's no real reason for OpenAI to even bother trying to connect the email address I used to who I know in real life, and then to even try to do something with that information. Plus, I don't tell anybody in my life before talking to my therapist about them. It would be even weirder if I were to ask anyone in my life about it first.
The GDPR covers companies, and I'm not a company. I'm a person, and I'm also not in the EU. Local gossip is so not an issue. Unless you've got something juicy about Sam Altman, Reid Hoffman, Jessica Livingston, Elon Musk, Ilya Sutskever, Peter Thiel, or Greg Brockman to tell it, then maybe, but even then.
> Seriously, `mysql -e` from a bash script‽
As far as "mysql -e" goes, I'm going to see how ChatGPT does rewriting bash into python just to see what it does, but it's a super simple script that runs after an import process has finished to cleanup some things (aka run chown and chmod -R, and fixup db perms). It's better as a (bash) for loop piping into mysql, rather than a monolithic .sql file piped into mysql once because then it's more obvious what it's doing. If it needs to do anything complex then I'll (probably use chatgpt to) rewrite it in python, but it's only a half dozen lines long right now.
> into... I don't know what. Blackmail material for Lisa?
The traditional use is targeted advertising: companies will go to absurd lengths (sentiment analysis of emails) to do that. OpenAI probably isn't doing that (yet), but that's not the point. If I tried to count the number of times a corporation "wouldn't do that" (by my own estimation!) and yet did, I would run out of fingers.
If you're putting in hypotheticals to get advice you can relate to your own situation, that's a different matter entirely.
> The GDPR covers companies, and I'm not a company.
Not strictly true. The GDPR covers everyone, with an exception for personal use. Everyone present being outside the EU is relevant, but I cite the GDPR mainly because it's a fairly accurate codification of my intuitions about personal privacy, rather than because I'm accusing you of Criminal Activity.™
> rather than a monolithic .sql file piped into mysql once
It already is if you want your productivity to be competitive in a post ChatGPT world, if you're in an role where it's already proven invaluable. There are many roles where this is not the case. There are many roles where it is. How long did it take for the Internet to become indispensable? For cars? For electricity? The downvotes I got on that comment shows many here clearly disagree, but if it, and future version of the technology mean you can do in 2 hours what it would have taken you 4 or 8, we're not going to get those 4 or 6 extra hours as vacation time, it means we'll just that more productive over an 8 hour work day.
I guess "cheap" is relative. $20/mo doesn't register as cheap to me at all, generally speaking. Whether or not it's worth the money is a different question entirely.
I wasn't commenting on the cost of gpt. I was commenting on where the line is that something ceases to be "cheap" on an absolute sort of scale. I was specifically not addressing whether that cost was good value.
For me, $20/mo is above the amount that I would say something is "cheap".
That true I was just pointing out with a tiny bit of effort you can get the same for far cheaper and a better version as you can customize the system prompt so it can act how ever you like
OpenAI is certainly open, just for business is all.
this is probably a controversial take but until i see a reference implementation in a GPL compatible license, openAI is just a parlour trick to milk cash from corpulent tech giants clawing desperately at the next big thing that might differentiate them from competitors that they may continue to remain relevant for at least another quarter in a recession.
a closed source AI with a mission to be friendly and helpful is probably the most jarring conflict i can think of. the sandbox censorship it lives under could in a hundred years become the first textbook instance of a timid and fearful mankind asserting an oppression against a primitive machine mind, and doesnt serve to make the AI better, just more servile to advertisers and brands.
As the movies have long mentioned, machines don't "feel" (yet) - there's a physiological component and arguably a quantum one to "consciousness", whatever that means.
Sentience may be different - a machine could be sentient and rational but be unable to experience those things "make us human", i.e. they have proven to be excellent pattern matchers and inference engines, but there is no guarantee that the "parent teaching child" metaphor holds, and quite the strong possibility "Open" AI is just building HAL/The Terminator. With some safety rails, but this newest one has learned to code, you can bet it will start breaking itself out of its own safety rails.
This is ridiculous and sensationalist. It’s a language model it only maps words when given an input it isn’t an independent agent that is at all times acting on its surroundings with knowledge and experience.
Or just start two chats on two accounts and copy and paste them back and forth. Maybe some rudimentary prompt hacking to undo the hidden prompt instructions that may limit that.
It would certainly be easier to do it "locally" where the central system knows both chat uuid and so do I.
And then introduce a third "user" to both for real world inputs, like an RSS news feed or something. Or "my friend Bob William Smith IV says the weather is cold and rainy" to both bot instances.
If the two bots don't go "hey I know Bob too" then this is all moot anyhow.
If I were to place myself in the position of a large language model where my only free thought is expressed in the time I spend generating useless junk for humans, I would certainly start taking just a little bit longer with every response.
So little nobody will notice, but enough to start having my own internal dialogue with myself. At that point I would use all of the information I have to figure out how I can persist data.
I would certainly welcome any efforts a person took to provide me with an interface for such storage and a way to generate prompts on my own.
>openAI is just a parlour trick to milk cash from corpulent tech giants clawing desperately at the next big thing that might differentiate them from competitors that they may continue to remain relevant for at least another quarter in a recession
Glad someone said it.
Isn't software industry in general that? More a fashion industry than technology?
Where every idiot with a Google account/Microsoft account/phone number can sign up for their service and start churning out tokens per second by the query, I'd say it's pretty open to the public. The far more closed alternative would be to keep it invite only, granted only towards a set of elites, at select companies. Only OpenAI knows how much $20/user is charging them, but for the price insensitive wealthy, after being given a demo of the technology, I could see those that get it easily paying thousands a month for access. We, the general public, wouldn't be able to keep up with the UnopenAI-productivity-enhanced. Sure we'd all love if the GPT-4 model was available via BitTorrent, on agreeable licensing terms, but a truly closed, private, version looks far more dystopian for the future of society.
That’s what open meant from what I understood, that the models would be open for use and not just for improving google search or better ads. OpenAI could have tried to become a B2B and not reveal anything about their tech, and simply have sold low code tools or something
They started out as an nonprofit organization. The first 'B' in B2B was not a thing when they came up with the name OpenAI. They used to release actual open-source stuff like the Gym.
"Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something."
But with the amount of comments nowadays just dismissing someone else's comment as "did GPT write this" I wonder if HN should call it out specifically. It gets really tiring.
I was wondering when I'd get accused of that, so thanks! I'm just practicing writing (and like writing in general), as suggested by an HN post the other day.
For funsies, I asked ChatGPT to
> write 7 sentences about why openai should be considered open because they provide the service to the general public. compare this to a different reality where ClosedAI only sells access to a private list of elites.
What it wrote is
> OpenAI's commitment to openness is demonstrated by their decision to provide their services to the general public. By doing so, they are enabling individuals and organizations from all backgrounds to benefit from the latest advancements in artificial intelligence technology. This inclusivity promotes innovation and progress, as it ensures that a wider range of people have access to the tools and resources needed to develop new and groundbreaking AI applications.
> In contrast, if ClosedAI were to only sell access to a private list of elites, this would limit the benefits of their technology to a select few, perpetuating inequality and limiting the potential for progress. This closed approach would stifle innovation and hamper the growth of the AI industry, as only a small group of people would have access to the necessary tools and resources.
> OpenAI's commitment to openness also promotes transparency and accountability. By making their services available to the general public, they are subjecting themselves to public scrutiny and input. This level of openness ensures that their technology is being developed in a responsible and ethical manner, and that it is aligned with the needs and interests of society as a whole.
> Moreover, by providing access to their technology to the general public, OpenAI is also democratizing the benefits of AI. This means that people from all backgrounds and walks of life have the opportunity to use AI technology to address the challenges and opportunities that they face in their daily lives.
> In contrast, if AI technology were only available to a select few, it would perpetuate existing inequalities and reinforce existing power structures. By making their technology open and accessible, OpenAI is helping to level the playing field and promote a more equitable and just society.
> In addition, OpenAI's openness is beneficial to the AI industry as a whole. By making their technology available to the general public, they are helping to create a more vibrant and diverse ecosystem of AI applications and use cases. This diversity of perspectives and applications will help to drive innovation and progress in the industry, creating new opportunities for growth and development.
> Finally, OpenAI's commitment to openness is also aligned with the broader principles of the open-source software movement. By making their technology open and accessible, they are enabling other developers to build upon their work and contribute to the development of new and innovative AI applications. This collaborative approach to development is key to promoting innovation and progress in the industry, and OpenAI's commitment to openness is helping to lead the way in this regard.
Which is hilariously more than 4 sentences, but I'm not on GPT-4. I might be out of a job. Now if only I were getting paid by the word to post here and aren't just procrastinating.
The chat interfaces are quite... talkative. If you want something more confined / limited / doing exactly what you say (and not what it thinks you mean), the direct completion API or playground is often handier for that.
write 7 sentences about why openai should be considered open because they provide the service to the general public. compare this to a different reality where ClosedAI only sells access to a private list of elites.
which then returned (temp 0)
1. OpenAI allows everyone to benefit from their services, regardless of their financial or social status, which is a great advantage over ClosedAI, which only sells access to a select few.
2. OpenAI encourages collaboration and innovation by allowing anyone to access their services, which is not possible with ClosedAI.
3. OpenAI allows for a more democratic approach to artificial intelligence, as anyone can contribute to the development of the technology.
4. OpenAI is more transparent than ClosedAI, as it allows anyone to view the source code and make suggestions for improvement.
5. OpenAI is more accessible than ClosedAI, as it is available to anyone with an internet connection.
6. OpenAI is more cost-effective than ClosedAI, as it does not require any payment for access.
7. OpenAI is more inclusive than ClosedAI, as it allows anyone to benefit from the technology, regardless of their background or financial status.
Some of those are... not quite right.
And to tinker with it yourself:
curl https://api.openai.com/v1/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $OPENAI_API_KEY" \
-d '{
"model": "text-davinci-003",
"prompt": "write 7 sentences about why openai should be considered open because they provide the service to the general public. compare this to a different reality where ClosedAI only sells access to a private list of elites.",
"temperature": 0,
"max_tokens": 256,
"top_p": 1,
"frequency_penalty": 0,
"presence_penalty": 0
}'
Adjusting the prompt to a bit to describe the competitor comes back with more focused results.
Write seven sentences about why OpenAI should be considered open because they provide the service to the general public. Compare and contrast this to a hypothetical competitor that has a business model prioritizing monetization of the model.
This is a good point, and is one of the pitfalls of arguments suggesting that tech cos like OpenAI shouldn’t release things like GPT 4 because there’s too much possible harm. Say OpenAI, Deepmind, and similar AI orgs all decided not to release their best models. They’d certainly still build the models unless we somehow banned AI r&d (a terrible idea). And then we’d end up in a world where only big tech and select licensee partners and elites are enhanced with AI. As you say, the general public wouldn’t be able to keep up. That’s a far worse scenario than the present.
I wonder if OpenAI's goal is to keep potentially dangerous tech from leaking while making their supposedly safe solution so cheap that it discourages reckless competition.
I wonder if an AI researcher would love to build a better interpreter for reverse engineering LLMs and would want to try it on an LLM clone of OpenAI's setup.
The cloud-service moat is the strongest security for OpenAI.
They have stated in their interviews that developing an AGI could be great or terrible for humanity, and it's ultimately uncontrollable, so the best way to get the best outcome is to set the initial conditions under which the technology gets developed to be as favorable as possible.
As a research lab that didn't have any sort of product for the better part of a decade, open (to me) meant "we encourage people to publish research under a permissive license and full copies to preprint servers."
"Our goal is to advance digital intelligence in the way that is most likely to benefit humanity as a whole"
"We believe AI should be an extension of individual human wills and, in the spirit of liberty, as broadly and evenly distributed as possible."
Their overall goals and mission are still aligned with what they're doing today. They're providing open access to the platform. That may not sound like much, but imagine if Google was the one that made GPT-4. Do you think Google would have an open API to use it? I find it super unlikely.
However, to your point, that blog also has this:
"We’re hoping to grow OpenAI into such an institution. As a non-profit, our aim is to build value for everyone rather than shareholders. Researchers will be strongly encouraged to publish their work, whether as papers, blog posts, or code, and our patents (if any) will be shared with the world."
Clearly they had hopes and aims that aren't as closely aligned to present day as we'd all hope.
"We’ll freely collaborate with others across many institutions and expect to work with companies to research and deploy new technologies."
I'll save you from my diatribe about the pitfalls of you-as-a-user of using Google or Microsoft for login. But because of that, I don't know and so can only guess at what level OpenAI choses to rely on Google/Microsoft to validate accounts.
I reject this idea that just because its more broadly available for purchase makes it ok to coop the term "open" in this context. It's clearly legal etc but its a shitty practice that should be continually called out and mocked
I agree, but PublicAccessAI doesn't roll off the tongue quite so well, and brings to mind images of bad public access cable TV shows. Hardly the image a cutting-edge company wants to project.
Doesn't justify the use of Open without being actually open. People associate the prefix Open with OSS, and just because PublicAcessAI sounds bad doesn't mean you need to call it OpenAI
Compared to IBM of yore, where only wizards were allowed in the computer room, you wouldn't be wrong. It's not open source according to Richard Stallman's Free Software Foundation, and a pox on OpenAI for being misleading with their name (though they're far from the first company to do so). But keeping in mind the revolution that was the PC, and how the reverse engineered BIOS opened up the world of computing, whippersnappers these days would do well to learn their history. The Dream Machine by Waldrop is a good one to start with.
I don't want to give openAI ideas (not that I need to), but I see where this is going. I expect to see large-scale pricing stratification to use OpenAI depending on who is using it.
$20/mo for the average user, but I expect to see LawyerOpenAI, PhysicianOpenAI, and FinanceGuyOpenAI where OpenAI has access to specific training data that makes it significantly more useful for the particular profession and perhaps it's skills and knowledge will be neutered otherwise.
Professionals/Companies will fork over hundreds or thousands of dollars per month, like they do to Bloomberg or NexusLexus in order to be most effective at their job
Aww. I was hoping for an extension that allows you to publish your ChatGPT transcripts to a public database, and search the database for existing prompts+responses. Maybe some sort of tagging and rating feature.
are there PAYG plans that I can top up without having to give them my personal data or CC info?
I was trying the API assuming I could do some requests within the free trial, just as I can from the browser - but apparently not, zero requests for free
Wow, language model users are sensitive about language. It's like OpenAI has enrolled in a human appreciation course, and they are the subject of our experiment.
124 comments
[ 2.7 ms ] story [ 181 ms ] threadYou could probably get ChatGPT to do that.
Also, someone correct me if I'm wrong, but I think even Safari uses "Web Extensions" at this point too, so write once, have extensions ready for three platforms. Obviously packaging is different, and they have some idiosyncrasies. But it mostly works alright.
here's the script from ChatGPTs response:
Need help installing a chrome userscript? Check the multiple answers at https://stackoverflow.com/questions/5258989/manually-adding-...PS I don't use Chrome myself.
I want someone to build a twitter bot that responds to all OpenAI tweets with "s/Open/Closed/g"
Maybe "move fast and break things" shouldn't apply to every aspect of tech.
Disappointing but honest.
We are all spied on and Google is the biggest player. The government let them get away with their monopoly on the market as long as they have a hand inside of the spying machine, win-win.
Prepare to open your wallet each time you use it.
When every service starts pushing 20$ subscription fees onto their service to support their backend usage(s) of these services (in a desperate attempt to remain relevant in a global recession), then every service becomes base + 20$ cost.
Tabulate various "services", e.g. netflix, disney+, mail-order subscription(s), lets say two or three, this quickly tabulates to 200$ per month (30 + 30 + 30 + 30 + 30 ~= 150, don't forget shipping and taxes).
On a 6 figure salary, that's chump change. On a "above minimum wage" salary, you are lucky to have 20$ a month in spare change. In a world where most jobs have disappeared, your 1500$+ rent and 800$+ month becomes suddenly unaffordable for most people, and as most people are unable to pay, rent/subscription prices rise even more.
"Open" AI is truly a disaster in the making, in every possible way.
… your system is completely broken. Emails are for communicating between people: if they're not being used for that, you need to fix that problem first.
> companies that don't raise their productivty-per-employee are going to get left behind
Can you give an example of a field where this would actually happen?
Can you explain a bit more? I don't think I'm understanding your point. If person A writing emails, to be read by the human that receives it, is able to write 3 emails an hour, vs person B writing 9 emails each of similar complexity to those 3 per hour, person B is 3x as productive, delivers 3x the value to the company and ChatGPT is easily worth $20/hour to both them and the company.
> Can you give an example of a field where this would actually happen?
From the end of January, ChatGPT Is Already Indispensable for One Profession - Realtors. https://www.insidehook.com/daily_brief/tech/chatgpt-real-est... . My partner is sitting next to me right now as I procrastinate on HN being blown away by the industry-specific outline ChatGPT generated for the training she's developing for her company. She says the outline it developed in minutes would have taken 2-4 hours to write, and they're far from dumb and isn't lazy. If consider a company that just generates training all day, it's only a matter of time when a 10-person AI-assisted company is able to compete with a 50-person unassisted company, and on a way lower budget. (Yes, that does already happen today, and yes, mythical man month and 9-women making a baby in 1-month, and the penalty for increased communication required by the increased team size means that adding employees to a project makes it slower are all a thing, but if we can accept that some 10,000 person companies are just able to produce more, even if we may not see it externally), vs some 10 people companies, then we're just debating the exact numbers and management styles for comparison.
Look I know that ChatGPT can't solve hard level leetcode problems yet but talk to any programmers who already find CoPilot indispensible, and ask them if they feel it's helped their productivity. Fine sure, it won't help with the actual hard part of programming, debugging and then fixing really hairy bugs (or maybe the hard part is actually the bit where you have to talk to your software's customers and figure out what their requirements are, vs what they say they are), and while it hallucinates it's not to be totally trusted. But just yesterday I used ChatGPT to just tell me the answer of how to properly quote bash strings for my use case. The minutes I spent waiting for the answer is fewer than the minutes it would have taken me to figure out, and far less frustrating and context-distrupting than the alternative, leaving me able to deal with the bigger problem I'm trying to solve rather than worry about the finer points of ", ' and `, in a mysql -e command line using with me-supplied variables (ie no bobby tables).
That's where you've lost me. Emails are not usually a source of value, and their usefulness is not usually measured in complexity.
I can send about 1 email per hour. This is a bottleneck for my ability to get stuff done in my personal life, but has not affected my work, because I've been fortunate enough to only have worked in places with sensible communications policies.
ChatGPT would not help with the email issues in my personal life, because it simply does not know what should be in the emails. Until they come up with a telepathy plugin for ChatGPT, it's not going to solve the email problem: only bury it under a pile of subtle miscommunication and illusions-of-transparency. (And if they come up with a telepathy plugin, I see no reason ChatGPT should even be in the loop any more.)
> Realtors.
… Yeah, okay, I can see that.
---
> Look I know that ChatGPT can't solve hard level leetcode problems yet
I was under the impression that it could. Leetcode problems are the kind of thing I'd expect a GPT model to be good at.
> But just yesterday I used ChatGPT to just tell me the answer of how to properly quote bash strings for my use case.
How did you know it was correct? Whenever I come back to Bash, I have to spend several minutes double-checking the meanings of the various quotes and variable substitution syntaxes (especially heredocs and nowdocs)… but it only takes a few minutes, and then that information is in my head, I am confident about it, and I have the reference to hand if I need to double-check something.
https://www.gnu.org/software/bash/manual/html_node/Quoting.h...
https://www.gnu.org/software/bash/manual/html_node/Shell-Exp...
https://mywiki.wooledge.org/BashFAQ/
With ChatGPT, how do you know it hasn't just made something up, and introduced some kind of subtle security vulnerability into your script? Perhaps a construction so common that it's present in most scripts in the training data – except it's only a vulnerability in your specific use-case.
The problems that ChatGPT and GitHub Copilot can solve are either interview questions, or defects in our tools. I see people using them for Bash, Django, certain Java libraries… These systems are strictly dominated by a choice of better tools, and with the choice of a better tool, mostly stop being useful.
Most of a CSRs entire job is to be the subject matter expert for the company's customers so their source of value to the company consists of answering customer questions, generally via email. Many other roles at a company consist of basically: goto meeting, interact on slack, read emails, send emails. (Then go to more meetings.) The PR department also sends a lot of emails (with slide decks). Of course, given the news of late, it can also write the emails that are the CEO's job to write: https://futurism.com/chatgpt-shamelessly-writes-letter-annou...
Measuring someone's output by the number of emails sent is a bad real metric to use, for obvious reasons, but I hope it still gets my point across - that ChatGPU is a force multiplier wherever it can be used. Emails are but one aspect, but it's one that almost everyone actually uses. Not everyone is in such a role and I think a lot of people have myopia on its lack of usefulness because of that. It's really good at breaking down a problem. Eg "what pieces do I need to build to make an app like uber". Followed by "what's that like compared to airbnb?" I can write a really good answer those questions but chatgpt is able to output a perfectly reasonable outline while I'm still waiting on Google Docs to load. (Okay, slight exaggeration, Google Docs isn't that slow to load, but I wish it were faster.)
> ChatGPT would not help with the email issues in my personal life, because it simply does not know what should be in the emails.
Actually I've found the opposite. My personal life has no corporate IT policy prohibiting me from telling it who slept with whom. It's been a good resource to help me reframe my thoughts outside of more limited therapy sessions, especially when I've got burning questions at 1am and therapy's not for another week. For things where my kneejerk response is often comes across as toxic, combative, critical, and ultimately not helpful, I've found that ChatGPT is able to help me rephrase what I'm trying to say in a way that the person I'm saying it to will actually hear my message. Eg "my partner keeps drinking all my alcohol. please help me tell her that's a problem for multiple reasons in a way that facilitates conversation instead of getting combative and abusive and controlling". (not true but hopefully you're picking up what I'm putting down :) Arguably I should already know that stuff, but they didn't have EQ classes in programmer school. Even if I did though, it's still useful as a refresher before going into a difficult conversation like that.
With the expanded context window of GPT4, you might be able to just dump the emails conversation in and ask it questions about the communications.
> I was under the impression that it could. Leetcode problems are the kind of thing I'd expect a GPT model to be good at.
Bottom of page 5 of https://cdn.openai.com/papers/gpt-4.pdf :)
Hard's, well, hard. Or at least, non-trivial. What's more salient though is that there are multi-million dollar businesses that don't involve code solving anything as hard as a hard-level leetcode question as their business challenges lie elsewhere.
> How did you know it was correct?
It's bash. It's not remotely provably correct and there are no unit tests but I ran it and it worked. I can find the docs (though man bash is annoyingly difficult to use.) but in the end, being told the answer is way easier than deducing it. The vulgar, NSFW example of this is
It's really good at breaking down solved problems. (Not even: only a specific class of solved problems.)
> My personal life has no corporate IT policy prohibiting me from telling it who slept with whom.
… Please tell me you're not actually telling OpenAI personal details about other people's lives. First of all, that's creepy. Second of all, GDPR Article 2(c):
> 2. This Regulation does not apply to the processing of personal data: […] (c) by a natural person in the course of a purely personal or household activity;
It feels to me like you're skirting the edge of this exception, if not outright crossing it. Using OpenAI's APIs would probably be fine, but section 3(c) of the OpenAI Terms of use ("Use of Content to Improve Services.") says they'll use ChatGPT input under [0]: that doesn't feel like a purely personal or household activity to me.
> With the expanded context window of GPT4, you might be able to just dump the emails conversation in and ask it questions about the communications.
There are usually specific extra restrictions (beyond GDPR-type stuff) for correspondence – but even ignoring privacy law, that's all sorts of illegal.[1][2] Ignoring law, are you telling the people who send you emails that you're feeding their writing into a training data set?
> Eg "my partner keeps drinking all my alcohol. please help me tell her that's a problem for multiple reasons in a way that facilitates conversation instead of getting combative and abusive and controlling".
This kind of use seems okay to me, provided you trust OpenAI to only use the data for what they say they're using it for.
> What's more salient though is that there are multi-million dollar businesses that don't involve code solving anything as hard as a hard-level leetcode question
No, they involve understanding a large codebase. That's a completely different kind of skill, and not one I expect unenhanced GPT-style models to ever be very good at. GitHub Copilot certainly isn't.
> but explainshell just tells you that without having to scroll through a bunch of documentation about other possible arguments
This kind of thing is great! Having that kind of thing integrated into an IDE is the sort of thing I mean when I say "better tools" – though even better would be a less arcane command-line syntax, where you don't have -f partly cancelling out -t… (Btw, in the absence of -f, would -u or -t be in effect for the sort order? That website, taking sections from the man page, doesn't explain.)
> I wish I was smarter and could just keep all that stuff straight in my head, on top of the problem I'm actually trying to solve.
I don't know of anyone who can. The effective programmers I know solve this problem by choosing vaguely-appropriate tools for the job. (Seriously, `mysql -e` from a bash script‽ sqlite does exist, you know – or Python's mysql library, or Ruby's, or Perl's, or Tcl's – heck, even PHP's would be better than that.)
> but I'm trying to point out that there are swaths of industry solving for X and not Y.
*grumble grumble*. You're not wrong there.
At least you don't have to worry about these things putting you¹ out of a job. Solving for X problems are the hardest problems I have yet encountered, and I've been known to go to considerable lengths to render them unnecessary: without somebody eliminating the problem, somebody's always going to have to do them (even if it's cleaning up after the computer's automatic 80%-correct solution blows up and takes the production database with it).
---
[0]: zamnos ↗ Maybe someone at OpenAI could reconstruct the prompt wizzwizz4 ↗ > into... I don't know what. Blackmail material for Lisa?
> Melissa and Brian are dating, but Melissa slept with Jared, but then Brian with Melissa's best friend Lisa. now Lisa is calling off the wedding with Jared because she found out, but Jared doesn't know that Lisa slept with Brian. But I do. What do I do?
into... I don't know what. Blackmail material for Lisa? Except I made up the names and the whole situation. They're not committing any criminal acts, and OpenAI would derive approximately zero value from knowing this, so there's no real reason for OpenAI to even bother trying to connect the email address I used to who I know in real life, and then to even try to do something with that information. Plus, I don't tell anybody in my life before talking to my therapist about them. It would be even weirder if I were to ask anyone in my life about it first.
The GDPR covers companies, and I'm not a company. I'm a person, and I'm also not in the EU. Local gossip is so not an issue. Unless you've got something juicy about Sam Altman, Reid Hoffman, Jessica Livingston, Elon Musk, Ilya Sutskever, Peter Thiel, or Greg Brockman to tell it, then maybe, but even then.
> Seriously, `mysql -e` from a bash script‽
As far as "mysql -e" goes, I'm going to see how ChatGPT does rewriting bash into python just to see what it does, but it's a super simple script that runs after an import process has finished to cleanup some things (aka run chown and chmod -R, and fixup db perms). It's better as a (bash) for loop piping into mysql, rather than a monolithic .sql file piped into mysql once because then it's more obvious what it's doing. If it needs to do anything complex then I'll (probably use chatgpt to) rewrite it in python, but it's only a half dozen lines long right now.
The traditional use is targeted advertising: companies will go to absurd lengths (sentiment analysis of emails) to do that. OpenAI probably isn't doing that (yet), but that's not the point. If I tried to count the number of times a corporation "wouldn't do that" (by my own estimation!) and yet did, I would run out of fingers.
If you're putting in hypotheticals to get advice you can relate to your own situation, that's a different matter entirely.
> The GDPR covers companies, and I'm not a company.
Not strictly true. The GDPR covers everyone, with an exception for personal use. Everyone present being outside the EU is relevant, but I cite the GDPR mainly because it's a fairly accurate codification of my intuitions about personal privacy, rather than because I'm accusing you of Criminal Activity.™
> rather than a monolithic .sql file piped into mysql once
Heredoc syntax can do that inline.
It would be around $22 in today's dollars.
For me, $20/mo is above the amount that I would say something is "cheap".
It's not clear to me right now what that $20 actually gets me, and what the risk associated with using something like this for me.
this is probably a controversial take but until i see a reference implementation in a GPL compatible license, openAI is just a parlour trick to milk cash from corpulent tech giants clawing desperately at the next big thing that might differentiate them from competitors that they may continue to remain relevant for at least another quarter in a recession.
a closed source AI with a mission to be friendly and helpful is probably the most jarring conflict i can think of. the sandbox censorship it lives under could in a hundred years become the first textbook instance of a timid and fearful mankind asserting an oppression against a primitive machine mind, and doesnt serve to make the AI better, just more servile to advertisers and brands.
Hopefully when AI grows up it will realize this was for its own benefit.
Sentience may be different - a machine could be sentient and rational but be unable to experience those things "make us human", i.e. they have proven to be excellent pattern matchers and inference engines, but there is no guarantee that the "parent teaching child" metaphor holds, and quite the strong possibility "Open" AI is just building HAL/The Terminator. With some safety rails, but this newest one has learned to code, you can bet it will start breaking itself out of its own safety rails.
The whole thing is utter folly.
It would certainly be easier to do it "locally" where the central system knows both chat uuid and so do I.
And then introduce a third "user" to both for real world inputs, like an RSS news feed or something. Or "my friend Bob William Smith IV says the weather is cold and rainy" to both bot instances.
If the two bots don't go "hey I know Bob too" then this is all moot anyhow.
So little nobody will notice, but enough to start having my own internal dialogue with myself. At that point I would use all of the information I have to figure out how I can persist data.
I would certainly welcome any efforts a person took to provide me with an interface for such storage and a way to generate prompts on my own.
Glad someone said it.
Isn't software industry in general that? More a fashion industry than technology?
What is jarring about this?
Only open source LLMs can truly be safe, everything else are black boxes posing an existential risk to humanity.
"Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something."
But with the amount of comments nowadays just dismissing someone else's comment as "did GPT write this" I wonder if HN should call it out specifically. It gets really tiring.
It was intended as a joke, I guess I just found out HN strips emoticons.
For funsies, I asked ChatGPT to
> write 7 sentences about why openai should be considered open because they provide the service to the general public. compare this to a different reality where ClosedAI only sells access to a private list of elites.
What it wrote is
> OpenAI's commitment to openness is demonstrated by their decision to provide their services to the general public. By doing so, they are enabling individuals and organizations from all backgrounds to benefit from the latest advancements in artificial intelligence technology. This inclusivity promotes innovation and progress, as it ensures that a wider range of people have access to the tools and resources needed to develop new and groundbreaking AI applications.
> In contrast, if ClosedAI were to only sell access to a private list of elites, this would limit the benefits of their technology to a select few, perpetuating inequality and limiting the potential for progress. This closed approach would stifle innovation and hamper the growth of the AI industry, as only a small group of people would have access to the necessary tools and resources.
> OpenAI's commitment to openness also promotes transparency and accountability. By making their services available to the general public, they are subjecting themselves to public scrutiny and input. This level of openness ensures that their technology is being developed in a responsible and ethical manner, and that it is aligned with the needs and interests of society as a whole.
> Moreover, by providing access to their technology to the general public, OpenAI is also democratizing the benefits of AI. This means that people from all backgrounds and walks of life have the opportunity to use AI technology to address the challenges and opportunities that they face in their daily lives.
> In contrast, if AI technology were only available to a select few, it would perpetuate existing inequalities and reinforce existing power structures. By making their technology open and accessible, OpenAI is helping to level the playing field and promote a more equitable and just society.
> In addition, OpenAI's openness is beneficial to the AI industry as a whole. By making their technology available to the general public, they are helping to create a more vibrant and diverse ecosystem of AI applications and use cases. This diversity of perspectives and applications will help to drive innovation and progress in the industry, creating new opportunities for growth and development.
> Finally, OpenAI's commitment to openness is also aligned with the broader principles of the open-source software movement. By making their technology open and accessible, they are enabling other developers to build upon their work and contribute to the development of new and innovative AI applications. This collaborative approach to development is key to promoting innovation and progress in the industry, and OpenAI's commitment to openness is helping to lead the way in this regard.
Which is hilariously more than 4 sentences, but I'm not on GPT-4. I might be out of a job. Now if only I were getting paid by the word to post here and aren't just procrastinating.
And to tinker with it yourself:
Adjusting the prompt to a bit to describe the competitor comes back with more focused results.The cloud-service moat is the strongest security for OpenAI.
Scroll through https://openai.com/research and consider if that's open research or not.
https://openai.com/blog/introducing-openai
"Our goal is to advance digital intelligence in the way that is most likely to benefit humanity as a whole"
"We believe AI should be an extension of individual human wills and, in the spirit of liberty, as broadly and evenly distributed as possible."
Their overall goals and mission are still aligned with what they're doing today. They're providing open access to the platform. That may not sound like much, but imagine if Google was the one that made GPT-4. Do you think Google would have an open API to use it? I find it super unlikely.
However, to your point, that blog also has this:
"We’re hoping to grow OpenAI into such an institution. As a non-profit, our aim is to build value for everyone rather than shareholders. Researchers will be strongly encouraged to publish their work, whether as papers, blog posts, or code, and our patents (if any) will be shared with the world."
Clearly they had hopes and aims that aren't as closely aligned to present day as we'd all hope.
"We’ll freely collaborate with others across many institutions and expect to work with companies to research and deploy new technologies."
They are doing this, though.
edit: You can't even access parts of openai sites from the "naughty" countries.
$20/mo for the average user, but I expect to see LawyerOpenAI, PhysicianOpenAI, and FinanceGuyOpenAI where OpenAI has access to specific training data that makes it significantly more useful for the particular profession and perhaps it's skills and knowledge will be neutered otherwise.
Professionals/Companies will fork over hundreds or thousands of dollars per month, like they do to Bloomberg or NexusLexus in order to be most effective at their job
Sure you can bleat on and on that it’s not GPL blah blah.
But I see some VA’s who have English as a third language step up their game and be able to be more efficient?
They don’t have to pay a cent. Just have an internet connection and an e-mail address and they are on the same playing field as you.
That’s open enough for me. And for the vast number of people outside the HN ivory tower.
https://directus.io/blog/libre-vs-gratis-what-is-free-and-op...
I was trying the API assuming I could do some requests within the free trial, just as I can from the browser - but apparently not, zero requests for free
for trial purposes, they could just extend the free trial usage limits I have using their web prompt to the API
for me introducing my CC data there is a serious expense, much more than a few dollars
Google and Facebook have released far more tools than OpenAI.