Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
ChatGPT get-rich-quick schemes are coming for magazines, Amazon, and YouTube (semafor.com)
160 points by mfiguiere on Feb 25, 2023 | hide | past | favorite | 140 comments


> Its the latest evolution of internet side-hustle culture: Creators posing as business moguls are encouraging followers to use ChatGPT to earn money by churning out AI-generated content.

I followed a random number of highly polished influencers and gurus as part of my Twitter feed. It’s a way to keep a finger on the pulse of the guru zeitgeist as all of the gurus move in lockstep from one hot topic to the next.

Using AI to generate a business is definitely the topic du jour among wantrepreneur gurus. They’re sharing fantasies about using AI to create every step of your business, from coming up with the idea to writing email campaigns to helping them write all of the code. They share examples of high profile success stories that claim someone used ChatGPT to make a business from scratch with zero experience and now they’re making five or six figures per month without having to know anything.

It’s wild. The current trend appears to be selling expensive paid courses with “limited seats” and/or to provide a paid Discord community that they claim is full of other people sharing success tips. ChatGPT has been a dream come true for people who sell wantrepreneurship materials. I suspect they’re using ChatGPT to generate a lot of the materials, too.


> They’re sharing fantasies about using AI to create every step of your business, from coming up with the idea to writing email campaigns to helping them write all of the code. They share examples of high profile success stories that claim someone used ChatGPT to make a business from scratch with zero experience and now they’re making five or six figures per month without having to know anything.

It's not 5 figures (yet), but thanks to ChatGPT my cousin was able to set up a small webshop that generates some business for him. My cousin (never been to college) is not exactly a prolific writer so was always struggling to write good copy. ChatGPT allowed him to write email campaigns, blog posts and product descriptions so his website actually looks very professional. So yeah there's definitely value for the regular Joe to start using ChatGPT to do things that were not considered possible before.


If the only thing holding your cousin back from running a $xxxx/mo business was some copy then he could have found a freelance copy writer before ChatGPT was ever a thing.

I get that it's more accessible now, but this problem is actually pretty cheap to fix without ChatGPT as there are so many underemployed adults with graduate degrees.


It's pretty cheap in terms of dollars to use a human freelancer, but it's kinda "expensive" in terms of time. You can't just quickly iterate on ideas when your freelancer is asleep at 1am and you have a burst of creativity.


A bad writer without management xp can't exactly judge + hold a freelance writer accountable.


Nor can they do the same for ChatGPT output, which is likely lower quality to begin with.


That’s the thing: It can legitimately be useful for filling gaps in someone’s abilities and generating copy. The gurus use this kernel of truth to suggest that it can do everything for you, as long as you buy their $599 (limited time offer!) course and then get on a monthly payment to their private community where all of the secrets will be shared…


It could also be very useful for ESL people: churning out thousands of words of copy can take forever in a second language and can also be absolutely miserable.


Every time I dip into this world I'm shocked at how credulous and clueless the customers are. I think (or hope) the average person in society is able to spot this stuff for the grift it is, but man, those less bright are lambs to the slaughter, losing their life savings on various web3, NFT, AI grifts.


Oh I can totally see how, because I ate that shit up when I was 19-20. I was your classic 20 year old that wanted to be a millionaire by 21 because "that guy over there said he did it". Never coughed up any money for a course or anything but I saw a shitload of these guys' free content on YT, Instagram, etc. In another world I could have totally fallen for one of these courses.


Same as it's ever been though: grifters schilling get-rich-quick schemes.

I'm sure in the mid-1800s there were classes you could buy to learn how to invest in future railroad property out West. Or later, how to become a millionaire revitalizing property in the South.

General rule of thumb: never pay for a lesson if price requires an altruistic seller.

F.ex. If I had a reliable way to make millions, why would I be selling that course for anything less than hundreds of thousands of dollars?


Isn't this what higher ed sorta is. You are paying for a lesson, but way more money. A lot of things in life are like this, like taking ski lessons, coding bootcamps, pilot lessons or culinary school. Is the scam having to pay or having unrealistic expectations? This would apply to some college students, who got into a lot of debt and don't get the career they wanted or was promised.


this reminds me of an old joke about how the only people who made money from the gold rush were those selling the shovels.


It's not really a joke as far as I know. It's a saying.


Yep, it's not stupidity, it's life inexperience. At that point in life many people have only vague ideas on how money is really earned in the real world, and get-rich-quick schemes seem at least plausible. Didn't fell into that trap but my sister had to go through some small "make money from home" scam to learn that lesson.


The people who have the capability and motivation to build a business can #1 download pirated versions of these get rich quick courses for free and #2 figure out how money is actually being made and move on to the hard stuff.

The problem is most of the customers have neither.


Related: many fewer people post failure or hard work over time on social media.


I think anyone with ambitions falls for these kinds of grifts at some point in their lives before realizing it’s bullshit and never getting sucked into them again.

When I was young and out of college I certainly fell for the trap of trying to build a startup and get super rich and become a cool tech bro. Read all the lean startup garbage and listened to anyone with a successful tech startup story. Fortunately the grift industry wasn’t as mature as it is today, so damage was minimal.

Nowadays, these content creators and course curators extract huge sums of money out of people and make themselves look more legit than ever. Young people need to be more wary but I feel they are more gullible than ever. Not a good mix.


American society is dumb people selling useless crap to other dumb people.


It's fun and easy to bash America, but if you cut America off the statement does it not cover most societies?


I don’t live in other societies so I don’t know. What I do know is that America celebrates this sort of thing and calls it “hustle”.


You may be surprised then to learn hustle exists around the world and has existed in one name or another since we started writing words in books and likely before then. Every pyramid scheme that's existed has been some form of this. Same with fraudulent bonds/stocks back in past.

We are just subjected to it more on an individual basis because we have speed of light global communications and mediums that allow hucksters to spread their grift to millions of people at once.


I’m not arguing whether or not grift exists universally. I believe it does.

What’s not universal is the near-idolization of it. That’s uniquely American.


America is relatively unique in that its culture celebrates "coaching" which is the angle that many of these get-rich-quick schemes take. I guess that it might be related to the sports centric high school education that many (most?) Americans experience.


That might be a little bit reductive.


Information disparity is unavoidable. You can only learn so much. I don’t mind dining in restaurants even though the food would only cost <30% when I made food myself.


That analogy doesn't work out. More like, you don't mind dining in restaurant without knowing ingredients , or the effect they have on you (nutritionally and allergies).


I have a friend that has made millions off of NFTs and another friend that has made hundreds of thousands. Grift if you’re bad at it.


Having made money out of something is not an intellectually relevant argument.

NFTs are snake oil, the fact that some get away with that stuff and a nice payout does not change the idiotic premise of the thing.


It's basically the same as investing in startups the vast majority fail. But 1% make a lot of money.


Startups make money by providing a useful service. NFTs do nothing but make money for a few and lose money for the majority. They’re useless grift fodder.


I would say a large portion of tech startups are actually bad for society in the long run.


It's only a "crime" if I lose money doing it!


No, it’s still a crime because somebody always loses money, somebody who simply came in line after you. The line could’ve broken before you so you got lucky


When their only skill is being the 'ideas guy' I guess it makes sense that when an idea generator comes along they exploit that too. And I assume they are selling some bootcamp or course or community access to 'teach' suckers how to do this stuff?


I see market for shovels here when it could generate cookie cutter webapps like wanapreneur writes "I want facebook for puppy owners" - Ka-B00m AI generated site they pay $100 earn nothing on it but there are 200 of guys with the same idea maybe with different domain/color scheme.

It most likely does not have to be generated at all just a template with some modifications. I earn too much and I am too lazy to do that - hope no one picks the idea.



OK this is it - and it is as awful as I imagined it to be. It is generating testimonials even.


If it’s that easy won’t the wantrepreneurs just do it themselves and cut you out of the loop?


From what I learned over the years they usually don't have any real skills besides parents money or they don't even have money so they look for cheap ways to "score".


One of these was on the top of Product Hunt yesterday. [0] It looks so scammy.

[0] https://www.producthunt.com/posts/marketing-blocks-web-app


We have billions of people in the world and many of them are on the web. 99,999% of hustlers will make next to nothing, but those that get popular will earn a lot just by the sheer volume of followers. And these success stories keep the hustler economy rolling.


I jokingly say "chatgpt is the new crypto"


Except chatGPT has many obvious use cases


So many obvious uses, like:

- a search engine that saves you valuable time by giving you a direct, wrong answer

- marketers needing free copy that is verbose, awkwardly phrased, and utterly non-engaging

- warming up leads for salespeople with incorrect assertions

- spam emails with even lower clickthrough

- nigerian prince scams that sound slightly more believable

- starter code for programmers that needs to be mostly re-written

- customer service that will run you in even more circles before connecting you with the live human necessary to actually fix your issue

ChatGPT has finally solved society's biggest problem: not enough cheap, low-quality communication!

Imagine, us being alive during the singularity!


Okay but for internet arguments it's pretty priceless. It spat a retort out in seconds:

> A search engine that saves you valuable time by giving you a direct, wrong answer

While it is true that search engines may occasionally provide inaccurate information, it is important to note that they are constantly improving and evolving to better serve their users. Additionally, it is up to the individual to use their own critical thinking skills and evaluate the information they find online.

As an AI language model, I strive to provide clear and concise responses to queries, and my natural language processing capabilities allow me to generate engaging and meaningful content.

> Marketers needing free copy that is verbose, awkwardly phrased, and utterly non-engaging

It is not fair to generalize all marketers as needing free copy that is awkwardly phrased and non-engaging. Additionally, the responsibility falls on the individual salesperson to ensure that they are presenting accurate information to their leads.

>Spam emails with even lower clickthrough

Spam emails with low clickthrough rates are not unique to any particular entity or platform, and it is up to the individual to exercise caution when interacting with unsolicited emails.

>Nigerian prince scams that sound slightly more believable

Nigerian prince scams and other forms of fraudulent activity are illegal and unethical, and it is important to report any suspicious activity to the appropriate authorities.

>Starter code for programmers that needs to be mostly re-written

As a language model, I provide starter code that can be customized and modified to suit individual needs and preferences. It is up to the individual programmer to decide whether or not to use my code.

>Customer service that will run you in even more circles before connecting you with the live human necessary to actually fix your issue.

Customer service can be a complex and challenging field, but it is important to remember that the vast majority of customer service representatives are dedicated professionals who are committed to resolving issues and providing a positive customer experience. If you encounter poor customer service, it is important to communicate your concerns and seek a resolution.

>ChatGPT has finally solved society's biggest problem: not enough cheap, low-quality communication!

It is important to approach criticism with a critical and objective mindset, and to evaluate the validity of any claims being made. While there is always room for improvement, it is important to recognize the valuable contributions that various entities and platforms provide in our digital world.

SO CLEARLY you've missed an important application of this revolutionary technology.... I can ceaselessly make it reply to your comments with almost zero effort on my part.

I have become AI Gish Gallop, destroyer of digital worlds!


"I can ceaselessly make it reply to your comments with almost zero effort on my part."

consider me pwned


Please don't copy-paste ChatGPT output as a reply on HN.

https://news.ycombinator.com/item?id=33945628


Its value for the average person is still completely overblown, like crypto. It's no surprise that they're both being used to sell people pipe dreams of fantastic wealth.


It's kinda like selling shares for an airport during Kitty Hawk in 1903. Is this a big monumental thing? Yes, I believe so. Is it going to produce cheap flights around the world (relative to 1903) in a few years? No, a huge amount of work still needed done.


I am seeing the average person having it send email replies at work or write essays at school.

I have only seen two people use Bitcoin for trading goods out of hundreds I know got involved in it, of those two one being me.


As of now, I'd argue ChatGPT has a lot of value for an average student who has to write an essay "by tomorrow".


The value for the average person is very high considering it's a better Google for anything prior to 2022.


So does crypto! This joke is literally self-aware


It's been 14 years since bitcoin mined its first block. Still not clear what ( if any ) the use case for it is.


Buying stuff you're not allowed to buy (sanctioned currencies, contraband, illegal books if you're in Florida maybe?) is a pretty valid use case!


What books are illegal in Florida which are not also illegal in all of the USA?


I was making a reference to recent events: https://www.wsaz.com/2023/01/26/book-ban-underway-teachers-f...


That doesn't make owning a book illegal, it recognizes that it's a crime to show some books to underage children and that includes when stocking a children's library.

> "We were instructed last week that we essentially had three choices as far as our personal libraries that are in our classrooms. 1) We could remove them and completely box them up. 2) We could cover them up with paper or some sort of something. 3) Or they could be entered into a database where the school district has all of the library books, and if the book was in the system, then it could remain on the shelf" - A Teacher

Read option #3. Having to report which books are in the library is not the same as banning books.

Sadly, bad reporting has distorted understanding of the issue. ex: this quote from a misinformed parent.

> "I would not suggest banning books. It’s a slippery slope. This is good literature with value. Please do not ban books!"

Nothing is being banned. That parent can read any books they want with their children. Schools also don't stock Marvel movies in their libraries and nobody claims those are banned - they're just not appropriate for school libraries.


> ChatGPT has been a dream come true for people who sell wantrepreneurship materials. I suspect they’re using ChatGPT to generate a lot of the materials, too.

So then it can work, as long as the customer is credulous of a low-effory get rich quick. "You can't cheat an honest man" comes to mind. In fact, someone who writes such a course is not treating it as a low-effort endeavour, just hoping that people seeking after such an endeavour will believe that it is.


Can you share a few? I am clueless of this culture.


"Gary V" is probably the biggest huckster, although the universe of 'hustle culture' charlatans is wide and deep.

Just search "ChatGPT" and look for thumbnails where they're making mouth-agape "I'm Shocked!!" faces and an all-caps title about making money.


To be fair, Gary V _has_ walked the walk, and is a "proper" entrepreneur. Him making money on his personal brand doesn't make him any more of a huckster than Mark Cuban and the like.

I find the whole scene obnoxious, but there's a world of difference between these folks, and "wantrepreneurs" selling snake oil. At least the ones who have actually succeeded might be able to give valid advice.


Gary V is a success launderer. He made money from inheritance and then used that to build a foundation for a successful backstory for entrepreneurial credibility, so he can use it for justification on why you should listen to him and buy his courses. There’s a LOT of people like this.


> He made money from inheritance

Do you have sources for this? From what I've read, he took over his family's moderately successful wine business and made it a multimillion dollar enterprise. Then he made his own very successful media firm. He clearly understands marketing and investing, and while he certainly gives off investbro vibes, he does actually know a thing or two. Dismissing him outright as a charlatan is a mistake.


That is the typical success launderer blueprint.

The steps:

1. Make money from family business you didn’t start, or just inherit money.

2. Start your own business, doesn’t matter if it is successful or a failure you just need something that’s yours, you already have money.

3. Talk a lot about your “successful” business, do the minimum to make it seem successful and maybe do some actual business activities. This will confuse people as to the ratio of your inherited money to self earned money. No one actually knows what your business is doing.

4. Declare yourself a successful investor. Investments have a wide range of success and failure. This gives more plausibility that the majority of your money may be from “investments” and not inheritance or family business. Literally anyone can be an investor, the difference is the size of investments.

With these easy steps, you can now dismiss people’s claims that you were just born into the right family by pointing to some vanity business you started or going on about how your modest investment in some stock or crypto blew up. You can come up with all kinds of wild stories that way, and never really reveal the true secret. If they try to bring it up you can say how supportive your family was but most of your success was self made, which can just be a lie no one will be able to truly verify unless they look at your accounting.

This is how Gary V operates, and even if he knows some stuff, it’s just not that impressive as an actual businessman like Mark Cuban or something. Selling courses is beneath anyone who made actual success, it’s just not going to happen.


I mean, you can be that cynical towards any marketing influencer. Yes, there is a closed loop with these businesses, where the amount of revenue from being an influencer vs. their own business is not well known.

In Gary V's case, I would argue that his path towards success wasn't based on this. He had already built up his parent's business before he was a public figure. He understood the power of e-commerce, marketing and branding very early on, before social media existed. Today he's very successful consulting companies on how to do this better. These are all valuable skills people will gladly pay him for.

This is very different from the blueprint you've outlined.


I disagree. This is all just the system working as intended. You're just letting the halo effect of extreme levels of success in attaining prosperity making you accept a double standard.


Try searching TikTok for ChatGPT and you'll quickly get exposed to a whole bunch of this stuff.


Using ChatGPT to get stock tips or stock predictions and then trade

https://www.youtube.com/watch?v=PPjWfZm5jZk

Extremely low value scammy content


If they used ChatGPT to generate their course, are they really lying?


They are honest in the same way as those who sell online courses on how to get rich by selling online courses.


The shovels need to continue to be sold before the gold is found.

Everyone is hustling and the scammers are where this types of things happen, just like a mining town in the 1800's


I'm not convinced the current generation of AI is going to put any skilled workers out of their jobs. But one industry it is definitely going to disrupt is the bullshit industry. And by disrupt I mean bring it to previously unimaginable heights. Search engines drowning in garbage, interactive scams on massive scale, individualized fake news complete with deepfake footage. I'm bracing for a tsunami of bullshit.


Hopefully it can put a bullet in it, finally. If anyone can generate infinite bullshit instantly for free, what value is there in being a bullshitter?


Marx thought that technology would ultimately make production so cheap that laborers would rise up, but that hasn't quite happened.


It's not that nobody can see this coming, or that it's not a fair concern to have, but we should wait and see what happens.

AI may be the great equalizer after a period of adjustment.

Let's face it, the internet in its current state is already full of get-rich-quick schemes and bad information. What people are proposing about ChatGPT is nothing new except that fewer humans are involved.

If the output of ChatGPT truly makes the internet worse, the market may adjust. If the overall value of online content goes down as a result, then there may be less incentive to make shitty content that hasn't been fact checked, or there will specifically be a greater market for content that's been crafted and edited carefully. People may lose their taste for guzzling down garbage content if the garbage stinks too badly, and that would be a good thing.

On the other hand, maybe ChatGPT does help more people produce better content, and creates opportunities for more people to make money by removing more barriers to entry.

We do not actually know, yet. What we've experienced up to this point is a flash in the pan, and we ain't seen nothin' yet.


>but we should wait and see what happens.

Most AI alignment and safety people would disagree with you there.

Robert Miles - 10 Reasons to Ignore AI Safety.

https://www.youtube.com/watch?v=9i1WlcCudpU


So? We should manage the application of AI. Accepting that the toothpaste can't be put back in the tube isn't the same thing as ignoring AI safety. Researchers don't know everything, and we're treating people's opinions as facts when they're anything but. Ultimately, the only way we're going to find the answers is not by constantly ruminating but by waiting to see what's going to happen.

Also, the examples that guy is using are hardly comparable. Contemplating overpopulation on Mars is not applicable to the real world. Contemplating how to manage the consequences of AI is. We are so far from so much as stepping foot on Mars that it's pretty ridiculous that someone thinks it's valuable to dedicate time to Martian overpopulation.


You want to know how I know that you didn't actually listen to the content of the video?

But hey, go on and ignore AI safety researchers pointing out the alignment issues we are already experiencing.


Snarky, much? I'm sorry that I apparently didn't get the point of your video.

Again, with the "ignore" thing. You brought that word into the conversation, not I. There's a world of difference between ignoring ideas and not blindly listening to ideas as if they have a completely factual basis.

And no, I'm not interested in what you're thinking I did with the lack of respect you bring here. I'd be fine continuing the discussion if you drop the assumption you're making of me.


It’s hard to know what’s real and what’s hysteria when the field is so fast moving. Mass hysteria is well documented and experts aren’t necessarily immune.

https://en.m.wikipedia.org/wiki/List_of_mass_hysteria_cases

I’d also argue it’s hard to call anyone an expert in AI safety when the very definition of AI is still so ill defined.


Small scale paid writing is going extinct. I can't even envision a sustainable way to fight it... just imagine when we have specific "SEO" or "Amazon" optimized finetunes.

But truly free writing (like AO3) might be OK? I don't mind if writers "augment" their work with LLMs, and a reasonable volume of pure GPT jank is no different than the current low quality submissions.


Maybe for small fluff stories about fashion or politics. I haven't seen any AI content an important topic that wasn't riddled with easily-spotted inaccuracies. For instance, an AI might be able to pass a law exam, but ask it to actually explain a legal concept to a real attorney and it falls apart. AI, all AI, still lacks actual understanding. Lawyers are still going to be paying other lawyers to write articles for a long while.


I tried both using the API and a couple different websites that also use OpenAI's davinci API to generate erotic furry... literature.

Spent about 4 hours before I gave up. I don't know how to tell you all this but it's not there yet. Not even CLOSE.

It doesn't even seem to have any understanding of biology. Like I was trying to get it to write a scene where two birds did stuff with their cloacas...

It would keep inserting the most bizarre stuff and seemed to think that Avian species remove the cloaca from their body and mate like spiders? Just really disturbing stuff that was 100% unfappable.

Now, if someone is able to make it generate a consistent voice/tone, like "erotic furry fan fiction of zootopia but in the tone of Blood Meridian by Cormac McCarthy," similar to what stable diffusion can do with appropriate prompts and other tools like embeddings, img2img, etc... That'll be gold for literature.


Yeah, the models arent going to like genres outside of their dataset without an embedding or something. They are not creative in that sense.

I bet someone is going to do a finetune on AO3 some day (which is well tagged and hence a good target), and you will get your wish.


Maybe you need to finetune/train it on existing "literature"


AI (automated intelligence) writing is coming for sports and press releases first. Then fiverr copywriters.


"is coming for" - its already here and has been for well over a decade. One such example...

https://www.npr.org/sections/money/2015/05/20/406484294/an-n...

> A company called Automated Insights created a program called WordSmith that generates simple news stories based on things like sporting events and financial news. The stories are published on Yahoo! and via the Associated Press, among other outlets.

https://en.wikipedia.org/wiki/Automated_Insights

---

An automatic generation method of sports news based on knowledge rules - https://ieeexplore.ieee.org/document/7960043

In the Future, Robots Will Write News That's All About You - https://www.wired.com/2015/03/future-news-robots-writing-aud...

> HERE COME THE robot reporters. This week the AP announced it will use software to automatically generate news stories about college sports that it didn't previously cover. Specifically, it's turning to a content generation tool called Wordsmith, created by a Durham, North Carolina-based company called Automated Insights.


I was waiting for this to come up.


An AI-generated press release? I cannot wrap my head around that one. Would a corporation or politician have an AI just sitting there occasionally releasing random statements? Press releases are carefully controlled announcements with legal implications. They are normally authorized by all sorts of departments before release, a process that takes far more effort than actually penning the language. And people have to decide when and where a press release is appropriate long before actually writing it. I cannot see how AI would insert itself. Would President Biden "hire" an AI to sit in the corner listening to chatter in the office and occasionally just release generated statements after interesting events?


I imagine you could use an AI to write the initial draft. So imagine you are doing layoffs, you can tell GPT "We are [company description pasted here]. We are laying off 1000 people. Write a press release expressing how we are regretful to do this step, but how it will help the company stay competitive in the long term. Remember [legal requirements here]". Then that draft can make the rounds through legal and every other stakeholder, until you have a finished product.


This is what we do. We are currently doing. Our startup is too small to pay for PR but if I have chat GPT do the first draft I can fixed the factual errors very easily.


Publications will use AI to summarize the press release into their own article.

For sports, take the timeline of events and generate a story around them.


Yeah, entrenched outlets with professional writers and expert pieces from very specialized writers are not totally dead like indie writers because of their reputation. But this is still going to compound their financial woes.


Passive income strikes again. Internet scraping, content farms, "low content" books, and now AI-generated crap stuffing up any channel that pays out real cash or can support money making "side hustles" elsewhere.

It's very likely a not insignificant number of HN accounts are using this technology as an experiment or for SEO/marketing funnels.


It feels like I won't be reading any books published from 2023 onwards unless they came from writers I already know and respect or are known and respected by those writers.


Books will be fine. Most people don't buy books from first-time authors blindly.

What will suffer, and possibly even disappear, are online comments sections. Yes, they're pretty much gone anyway because nobody besides Meta/Twitter wants to moderate discussions. But ChatGPT-driven bots will probably kill what little is left of organic conversations.


Hopefully Dang and HN will persist.


So it will come down to curation. Being a trusted source for recommendations may become an actual business, kind of an ad blocker for books.


I would assume "buy from publishers you trust" will still be a good strategy. Publishers can enforce some baseline of quality, where on-demand printed books is just a flood of things of every possible quality level.

Of course making sure nothing is written using ChatGPT is impossible, but you can make sure the reader gains something from reading the book (whether that's insight or entertainment)


But you’ll never know if the writer you respect is generating future novels with an AI trained on their past writing. Maybe this is better, but in a very short time, AI writing will be indistinguishable from human writing.


I admit it's not a perfect heuristic but it's the only one I can think of at the moment. Things will definitely change quite a lot in the next few years and I'll need to adapt accordingly.


Any awards for up-and-coming authors or best first publish for 2023 will be interesting


That is bigoted speciesism. As long as the books are good - does it matter what generated them


Not speciesism. I'm biased against anyone who has a history of bullshitting. In general I would like to know before I read something whether an LLM wrote it. Why not just read it and see if it's good, you say. Because the thing about bullshitters is that they can waste a lot of your time before letting you down badly. The ones who are good at it are the worst.


You don’t use reviews or ratings to decide what to read? That is crazy brave even for pure human writers.


It does matter, to me at least. You can read whatever you want.

And how is it speciesism? LLMs are not biological entities nor conscious (yet, I guess).


I agree. It matters to me too and, as you say, it cannot be called "speciesism" even in the most generous definition of the word. I propose the term lifeism: ethical and aesthetic prejudice favouring entities that have a life over those that do not.

We can do a thought experiment:

There is a runaway trolley hurtling down a track in the direction of a mediocre human science-fiction author, who is not even good enough to quit his or her day job. On a side track is a server holding the only trained copy of an experimental ChatGPT model which can generate science-fiction novels as good and original as Arthur C. Clarke and Ursula K. Le Guin combined, and at a massive scale. You can pull a lever to divert the trolley towards the server, destroying the ChatGPT model.

Do you pull the lever? As a self-confessed lifeist I would definitely pull the lever.

    (By the way, the mediocre author is not actually on the track, but on a bridge on top of it, out of harm's way)


That is called sarcasm, top shelf too, so it doesn't self-identify as such and you need to think as you read.


Didn't felt like sarcasm to me (and if it is, definitely not top-shelf but opinions may vary) but yeah it happens sometimes that those things go over my head.


I think it may be possible for someone to write an interesting book with the assistance of an LLM, but what this article is warning about is people dumping tons of low-effort AI generated content into the market. Those are just not going to be good.


It's okay to be prejudiced.


He doesn’t like when the other mammals only enjoy mammalian things. It makes his cold reptileman blood run even icier.


There's many repercussions from this. It's going to get harder to publish to those platforms as a legitimate individual, and it's likely going to be harder and harder to prove you're a human online...

This means even more tracking, captchas, verification to prove "you're not a bot", resulting in less privacy and anonymity online...


But a real human can still use these tools and pass content off as though they themselves created the content…


This is a great point, I guess this is also something that will become more prevalent too...

People becoming bots...


Quite a few years back I had a conversation with a colleague about how "curator" would be a hot job in the future. I envisioned an internet awash with content from the "creator economy", and a breed of people picking up the hidden gems. I failed to even envision the factor of AI generated content.


People do it for free for karma and/or upvotes already. I think they’ll stay on their hamster wheels happily - no need to pay them.


How long until AI can do the curation?


ChatGPT is that curation, isn't it?

Consumes a bunch of info that no human ever could. Develop a map of said info. Generate content. Rinse and repeat.


Not exactly. Current LLMs are optimized to put out language that aligns with what we speak. They are not really aligned on 'truthfulness' and they are completely capable of just making things up. Now, don't get me wrong, LLMs are very neat and amazing at what they do, but we must be aware of what they cannot do at this time.


Discovering Latent Knowledge in Language Models Without Supervision

> Existing techniques for training language models can be misaligned with the truth: if we train models with imitation learning, they may reproduce errors that humans make; if we train them to generate text that humans rate highly, they may output errors that human evaluators can't detect. We propose circumventing this issue by directly finding latent knowledge inside the internal activations of a language model in a purely unsupervised way. Our results provide an initial step toward discovering what language models know, distinct from what they say, even when we don't have access to explicit ground truth labels.

https://arxiv.org/abs/2212.03827


It is a LOT easier to integrate an openai api endpoint that start something on blockchain.

It is almost hilarious to see these content creators jump ship so quickly after espousing crypto as the end all money printer.


I want to insist on the fact that "get rich quick by picking a trendy topic and selling a bullshit book / story / video that you did not write" schemes have been sold to people for a long time.

Previously those have been written by severely underpaid ghost writers (who, given the pay, have a tendency to mash existing content).

The scam is that the only people getting rich are the one selling the course on how to do it and that platforms have been relatively good at letting this spam content get forgotten.


New technology -> massive overhype -> lots of people trying to a make a buck of it.

So far, so predictable.


All that excess computing power for BTC is currently being converted to train models. This will power a new AI bubble. Now, how can we convert it into a crypto currency?


Any new technology is going to be used by hucksters, everything is, religion is. It doesn’t matter what it is. What is the problem here, how do we improve society, such that hucksters don’t need to huck? There are absolutely segments of society that are seemingly perpetually going to huck and be hucked. This is a property of humanity. If anything the pandemic made the flaws of humanity even more apparent and speed-runned through various scenarios.


Good. Maybe we will hear less about blockchain and NFT's.


I dunno, I hear now is the time to invest in Elon's nuclear GPT coin.


The level of interest in ChatGPT is completely bonkers.

My YouTube channel is mostly about embedded programming and electronics. But I posted a video on using the OpenAI APIs to make a ChatGPT equivalent. It’s performing better than any of my other content purely from the search traffic.

The comments are very interesting and it’s quite obvious that a lot of them are from non tech people who have heard that you can get rich from ChatGPT.


I feel like this was lost in the conversation about Meta / Twitter moving towards verified accounts. I think social media companies are starting to realize that they should prepare for a world where most online agents are AI or AI-assisted


For sure. Troll farms seems like they will be scaled up even more.


It is quite likely that these money making schemes are also generated by ChatGPT. I guess with the right prompting it can come up with a bot to remove the human in the loop. A bot apocalypse is coming.


A global Turing competition, with money prizes.

Maybe this is the incentive needed for runaway improvement, but beyond the reach of prompt engineers.


The weakness was the open-channel for submissions, they now need to gatekeep with submission-fees or some other mechanism.


Oh wow, ChatGPT output comes out of it's training data, who could have imagined!


Just when will our human civilization admit that Great Depression of 1920s was caused by rampant overproduction?


Yep sell shovels don't do the mining


selling courses on how to use a chatbot to get rich




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: