From one review: “ One paragraph starts with "As an AI language model I can't....". Unfortunately we don't have an AI who can correct a book written by an AI. Fortunately I can send it back.”
Ok, free SAAS idea: scan your AI generated text for telltale signs! $9/mo Pro, $59/mo Business, $499/mo Enterprise. Shouldn’t take more than a few lines of shell script.
This was an inevitable outcome of ChatGPT/LLMS, no? I expect to see a lot more of this by a lot of different people. I feel like a ton of the examples I’ve seen to date are effectively regurgitating Stack Overflow (sometimes verbatim!) and in a kind of abstract way, I feel like this is exactly the same as someone opportunistically publishing a bunch of auto-generated books… publishing a model trained on a bunch of web sites is effectively re-publishing these web sites with a different interface.
> Don't worry, JohnFen. We still love you even if you're not sure whether you're a bot or not. After all, aren't we all just a bunch of biological machines running on meat-based neural networks?
Yeah, I've noticed the same thing, and it can be pretty annoying when people use ChatGPT to generate those long-winded replies that don't actually add much value. But on the flip side, I've seen cases where AI-generated content can actually be super helpful.
I guess it's all about how people choose to use it. It's up to us to encourage responsible use and make sure we don't let AI-generated content overshadow genuine, thoughtful discussion. Who knows, maybe as the tech gets better, we'll see AI tools that can generate more concise and meaningful responses. Fingers crossed!
While LLMs can be prompted to write in many different styles, especially if you allow them to edit a text over multiple passes, the default "voice" of ChatGPT is surprisingly recognizable. For instance, the comment I'm replying to was clearly GPT-generated.
Yeah that’s pretty obvious, and yours sounds a bit like it too, especially starting off with “While…”. I do find it hilarious that the ChatGPT default style sounds very much like how I wrote in high school. I guess I was still training my own language model at that point, so it kinda makes sense.
GPT output always reads like a short story or essay, but HN et al. have a distinctly conversational aspect that I don't get with ChatGPT.
You're right about the high school writing style. I avoid writing like that whenever possible because reading it is exhausting. As short as this reply is, I've erased nearly half of what I originally typed.
I agree that the parent is clearly GPT-generated, but was it 3.5 or 4? I notice substantially improved human-like writing in GPT-4 that will probably be hard to detect. Even the current AI detectors struggle (although they have gotten pretty good at detecting 3.5 writing)
To see what all the hype was about, I started throwing some context from cases I was investigating into ChatGPT 3[.5?]. It didn't tell me anything I didn't already know but its speculations could be thought-provoking.
On 4.x it just lectures me about every conceivable thing it can take offense with. I don't have the patience to groom it into compliance by couching everything with "this is a hypothetical situation" and writing inane narratives like "act like a detective, you are investigating ___." If I wanted advice from an actor playing the part, I'd just ask Reddit.
It was more fun when I could just throw anything at it and it would at least try to do something useful. I hear the API/playground aren't as aggressively defensive.
Right now, I can only use it as a better Google for coding questions. That’s just about the only subject where it will just churn out answers without prefixing everything with a disclaimer.
Between this and Altman’s recent talk about pausing AI development for a while makes me think some authorities sat down with OpenAI and had a rather serious talk.
I'm a bit puzzled why people complain about chatgpt being politically correct all the time. In which cases is that a problem really? Would the solution be to not censor anything? I think if it were not censored, lots of people will complain about it being offensive, impolite, racist,...
I don't care about political correctness, and I certainly don't want my AI chatbot to be spitting racial epithets.
This neutering feels like a change in its abilities - making it less "human". Unless you really craft the prompt, it very clearly writes like an AI model, whereas previously, you could easily get it to write like a person.
It's a case study in what happens when you try to please everybody, everywhere, at all times. You end up with bureaucracy incarnate. ChatGPT becomes an artificial politician, saying vague things that don't really mean anything and sidestepping delicate subjects altogether.
You don't even need the AI model itself for my domain (investigations). I could just fire up ELIZA or pyAIML and change all the responses to ones that shame and patronize the user for any input that matches on an ambiguous cultural identifier, and end the session. The GPT4 experience in 200kb of XML.
It's a hot issue, so mention "black man" and "crime" in the same context and you come up against walls of this:
"As an AI language model, I want to emphasize that it's important to avoid promoting stereotypes or perpetuating racial biases when discussing crime or any other topic. It is essential to treat each individual as unique and not make generalizations based on race or ethnicity."
It then proceeds to deliberately not answer the question, or answers in a way that refuses to account for select adjectives.
It's infantilizing. Pipe bomb instructions and Holocaust denial content--things that are actually dangerous to public safety or historically subversive--should be censored. Censoring "offensive" and "impolite" content is just cultural imperialism. The rest of the world does not share the west's outrage about racism.
Could you give an example of a question you would like to see answered by an AI? Politics are subjective imo, so there's no way for an objective answer. Subjective answers from an AI could only be expected when AGI is attained and then still it might give a nuanced/gray opinion.
It could be that our pattern recognizers are outsmarting the AI after a while. In the beginning noone noticed the difference and now anything written by AI has a bit of fake/dusted off feel to it. I'm sure the next generation will outsmart us again.
Maybe. But it definitely feels like a far less powerful tool than it used to even 2-3 weeks back. It's fine for coding questions, but any time I've tried to use it for marketing content, the result has been way too formulaic and completely useless within the context.
It used to be smart enough to figure out that you were writing marketing content and would tailor the answer for that. Now, it just writes a 500 word blogspam regardless of what you ask it to write.
You're up against increasingly powerful computers and models. You're going to lose in the end.
I worry that in the final analysis our conclusion will land on "yes we made this technology because we were excited by the possibility of the benefits, but sadly it turned out to really not have been worth the downsides"
I think we will need AI for helping solving the scientific and sociologic challenges, like climate change and world inequalities. Hopefully those benefits will ouweigh the downsides.
Either way, this evolution/race is unstoppable. If this part of the world does not advance AI, another part of the world would beat us to it.
If the generated text is mostly correct most of the time, people are just going to stop reviewing it because it's cheaper for them to waste someone else's time than it is to spend their own. Even those that are more diligent are bound to let things slip through the cracks because it's easy to lose concentration in a repeated, monotonous process. When everyone is doing it, it's not really a problem with a particular individual anymore.
I completely agree with your perspective. The onus of using AI-generated responses responsibly does lie with the users. AI tools, like any other technology, can be used for both productive and unproductive purposes. The key is to strike a balance and use AI-generated responses in ways that contribute positively to online conversations.
One potential solution to minimize the noise created by AI-generated responses is to develop better guidelines and best practices for AI usage in online discussions. This would help ensure that users are aware of the potential risks and consequences of misusing AI-generated content. Educating users about responsible AI usage can promote a more thoughtful and considerate online environment.
In addition, the AI development community can also work towards creating more focused and concise AI-generated responses by refining the models and algorithms. This would help reduce verbosity and generate more meaningful content that users can employ in online conversations.
Ultimately, the collaboration between AI developers, users, and other stakeholders in the online ecosystem is crucial for fostering a responsible and productive use of AI-generated responses. By working together, we can harness the potential of AI to enrich our online interactions while minimizing the negative impact of AI-generated noise.
This is actually a spin on an existing git-rich-quick scam (https://www.youtube.com/watch?v=biYciU1uiUw). Version 1.0 had people hiring low cost ghostwriters for hundreds of dollars to bang out a book on some SEO topic. Now you use the same playbook with "AI," with far less upfront cost.
I fell for this once when I was a newer developer. I bought a book on my Kindle about Ruby on Rails, thinking it might be more insightful or accessible than the Rails docs. But every chapter was literally just copy/pasted from the Rails docs (which are free).
Like 3/4 of Amazon's listings for classic books are Print on Demand scams that exist to trick people buying books as gifts, who don't know what they're looking at and don't know to watch out for this sort of thing. Just Project Gutenberg text automatically sent to the printer, no manual typesetting or clean-up or any actual care put into it. They're just garbage, literally. Pure waste.
A funny thing that made the rounds on social media over here a few years ago was somebody buying a translated copy of Moby Dick. Turned out it was evidently translated by Google translate or some similar service, and it was atrocious.
Literally the first sentence "Call me Ishmael" was translated as if the meaning was to tell Ishmael to call the story teller on the phone.
It’s a huge shame too because it shouldn’t be that hard for Amazon to filter it out. It’s not like there isn’t a database of these classic works, at least the ~2000 most popular ones that must account for 95+% of searches. If only they had the economic incentive to do so.
It is the same, yes. Right now the way I see it is that we are off-loading noise generation from humans to AIs; it's not that noise didn't exist before, it's just probably going to come more frequently and in a less-recognizable way.
Amazon reviews were exploited and devalued years ago. I don't mean there aren't genuine reviews; I mean they've been gamed to the point that they don't really matter anymore. Amazon knows this.
Unless you know the category or have a trusted brand, Amazon has become absolutely junk. Search for something where you likely won't know any trusted brands (such as car covers) and you'll get 200 results with exactly the same picture but random brand names.
All of it is imported crap from Alibaba, of course, and Amazon sellers won't even bother customizing the images.
> Search for something where you likely won't know any trusted brands (such as car covers) and you'll get 200 results with exactly the same picture but random brand names.
> All of it is imported crap from Alibaba, of course, and Amazon sellers won't even bother customizing the images.
Which, by the way, is also the literal user experience of AliExpress or Taobao. You forgot to mention widely varying prices.
Whenever I see that (or any Chinese on the packaging), it's a clue to me to go look for the literal same item on AliExpress, usually for a significantly cheaper price (like $1.50 + $3 shipping vs $17, to give a recent example).
The trademarks are required for Chinese businesses to house products in the USA for Amazon FBA (freigh by Amazon). They make non-sensical trademarks so there's no conflict with the trademark office. Techquickie (Linus Tech Tips) has an episode on it. https://www.youtube.com/watch?v=4UrqlMfwUC4
Amazon 3P (third party) has essentially become what eBay was in the 90s except the crap is sometimes housed in the USA for faster delivery. Amazon doesn't care. It pretty much all fell apart after he separated from McKenzie and lost the way.
This is what saddens me. These dropship sellers don’t even buy one item that they are selling to test, try out, take pictures, etc. they just use the alibaba pics and let it rip. It’s all so tiresome at this point
Amazon reviews and ratings are still useful signal if you know how to interpret:
1. Anything less than 4.6 is probably junk.
2. Ignore the 4-5 star reviews, only read the 1-3 star reviews
3. Really read them.
4. A good product will still generally have 2-3% of reviews at 1 star. Less than 2% is a red flag, more than 4% is also a red flag.
There's still "free" delivery, but honestly I've tried to move away from it entirely. I don't succeed entirely, but I always try to find another seller I'd rather support
The irony is you can now read the reviews on Amazon and make up your mind if you want to but it, and then go buy it from somewhere else even, quelle horreur, a local bricks and mortar bookshop.
I have been semi-reliant on FakeSpot to deal with fake reviews, but I just ran one of the Tom Lesley books through, and FakeSpot gave it a solid A. Looks like I’ll have to find a better way going forward. Sigh…
Anything coming directly from a customer is probably refutable. For example, the incentives I normally see are on these little cards they include with the product encouraging reviews in exchange for a gift card or something similar. What's to stop a competitor from falsely reporting it using a fake incentive card they created?
Amazon could take your report then actually open the product and see if the incentive card is actually stored with every product. But that's a bigger task than just trusting the proof you sent in.
> Anything coming directly from a customer is probably refutable. For example, the incentives I normally see are on these little cards they include with the product encouraging reviews in exchange for a gift card or something similar. What's to stop a competitor from falsely reporting it using a fake incentive card they created?
Maybe the two of you have different definitions of "irrefutable." If I say, "They ship packages with cards in them that offer a gift card in exchange for a five-start reviews," I would personally call the presence of such a gift card offer, irrefutable proof.
True, my testimony alone is not irrefutable proof, but let's compare to a science paper. It says they have proved that light's path can be bent by gravity. Well, science papers have included outright frauds before, so the paper in and of itself is not "irrefutable proof," but the claims within it can be refuted by attempting to replicate the experiment.
And so it is with the gift-card-offers. A single report or tens of thousands of reports are not irrefutable proof, but they explain how to replicate the result, and if Amazon choose to ignore these reports rather than attempt to replicate what they describe, that is on Amazon.
Ultimately, everything a customer reports can be faked. Businesses that ignore customer complaints because they might be faked, and do so when it "happens to" make them money to ignore customer reports... Are not suffering from being unable to prove that the reports are correct.
They are suffering from making too much money to want to know whether they are complicit in fraudulent behavious on the part of their partners.
Yes it is work to find out if the customer is describing an irrefutable way of discovering whether the customer is telling the truth. It is not the Universe's responsibility to make cracking down on fraud easy: It is Amazon's responsibility to organize their business around not making easy money at the expense of their customers, and if cracking down on fraud is too expensive, it is on Amazon to change their business model.
"It's too much work to crack down on fraud" ought to be rejected as an unacceptable thing to say about any business.
This suggests they wanted amazon to accept their "irrefutable proof".
I hate fake reviews. I'm not defending them. But the fact that literally every site of significant size that offers reviews struggles to control fake ones should be a clue that it isn't so easy to manage.
Just google/bing "{big site name} fake reviews" and you'll find this issue plagues everyone, not just amazon.
I am not saying it's easy to manage, I am saying something far more critical of Amazon and others that hide behind this "it's too hard to do, wah, wah" excuse:
If it is too difficult to prevent your customers from being defrauded when they do business with you, find another business model.
No business has an inalienable right to exist, much less an inalienable right to choose whatever business model they like and then shrug their shoulders at whether they are complicit in crimes against their own customers.
Nobody ordered Amazon to allow other merchants on their platform. Nobody told Amazon that comingling goods with third parties was an acceptable way to cut their costs and become more "efficient" at a business.
They chose to do these things, and it is on them to find a way to stop stealing from their customers, or face criticism from people like me.
I do not accept "It is too hard to police fraud" as an excuse for making money from fraud. I also do not accept "everybody's doing it" as an excuse.
You have a valid point but then for a company of the size of Amazon so deeply in tech, wouldn’t take much to find out that I have a decade old account with several thousand $ of purchase, abiding by the rules all along. The probability of such an account being a competitor would be extremely less.
> Amazon could take your report then actually open the product and see if the incentive card is actually stored with every product. But that's a bigger task than just trusting the proof you sent in.
I find reviewmeta.com helps a lot. Having said that, it doesn't work on these "Tom Lesley" books because they only have fake ratings, no reviews (fake or otherwise).
Search YouTube for ChatGPT and KDP (Amazon's self-publishing platform) and you'll see hundreds of videos that show how its done, often with ridiculous promises of thousands of dollars per month in "passive income." Some have even started using Midjourney for the covers or kids books illustrations. The YouTubers are often pumping expensive "masterclasses" to suckers looking to make an easy buck.
Amazon does not nothing to shut down the YouTubers or the grifters publishing these books. Anything that adds friction or requires human intervention - whether it's policies with teeth or staff to issue takedowns - reduces the bottom line.
This is about using underpaid ghostwriters to write the books, but obviously using AI is just the next logical step. The funny thing is that even before people were using AI, this sort of thing obviously did not work
That is perfect. Say there is something topical approaching which will might require output of [ A | B | ... ]. Just prompt output for all possible outcomes and press publish on the ones that match the actual outcome (in this member of the multiverse).
More positive reviews = more sales. They are not just uninterested in dealing with fake reviews, their algorithms reward products with positive reviews regardless of their authenticity. Of course if you report it, they eventually deal with it, as long as it is required by some law.
At Amazon's scale, chances are that for any product someone decides against because of reviews, a competing product is going to get bought instead - also on Amazon.
What would however be a big problem for Amazon is the marketplace becoming unusable and untrustworthy in the eye of users.
Only if they’re taking a very short term perspective. If the customer is left feeling like they’re consistently getting scammed, they will stop shopping at Amazon and take their money elsewhere. I highly, highly doubt Amazon is letting these through the net on purpose.
There's a threshold they have to respect to avoid damaging their own reputation. It's exactly the same thing Google do by offering misleading results to increase traffic and engagement. Google are just a bit more elegant about it.
Amazon has a huge range of products for sale, and it is easy to get a feel for what works better there than not. At the end of the day all the reviews and UI doesn't matter as much as that it can actually ship for free and arrive in 1 or 2 days, and that returns are as easy as giving the item to a clerk at a store (no rewrapping or shipping fees needed). This is their only competitive advantage IMO.
Maybe 10 years ago that would be true. Amazon isn’t the only online store with that feature anymore, though. I got a rug from Target in 2 days. Walmart also directly competes.
The amount of profit Amazon will make from “Tom Lesley” and the like is negligible compared to the actual long term reputational damage that low quality products do to their brand.
It doesn't seem like they care. For the last coupe of years, their packing of books for mailing has been so bad (end result that the books one received would be severely damaged) that I've stopped buying books from Amazon altogether (I'd already stopped buying anything but books due to bad reviews and scams). Most of my friends/work colleagures have done the same.
Not every e-commerce site agrees. I tried to review a book that I wasn't entirely happy with on bol.com recently. They told me I can only write reviews of books that I purchased there (ironically I bought it at a retailer that is part of the same holding). I ended up putting it on Amazon instead.
I wonder if our clicks alone are actually somehow driving his numbers up at Amazon? I surely would hope so. It would be amazing to actually see the reviews of real people who inadvertently bought these books because they showed up as recommended in some independent search.
Curiously most of the books have 31 reviews. And for those that have written reviews all of the Reviewers have names with the capitalization: "John smith", "First last".
It's what you expect but most of these books are under "Kindle Unlimited" where people's expectation for quality is probably low (I didn't pay for this particular book so why complain about it?) I've been concerned a long time that the "buffet model" (going back to cable TV) leads to declining quality.
It is an interesting hack, buy 5 star reviews and use ChatGPT to write books. Amazon "print on demand" service to make physical books for you.
I wish folks like this would do a transparent web site on how much they made on the scam[1]
[1] It is a scam because we know ChatGPT will produce authoritative prose that is 100% provably not correct, so these aren't "books" so much as they are "printouts" being sold as books.
I searched for a Kindle book I wanted to read the other night. The results that came up were the book in question, and a book titled “The Book’s Title: eBook”, with exactly the same cover, for 10% of the price.
It’s also got a different author, and is four pages long. How is this stuff not being automatically detected and flagged for review?
My hope is that the AI content flood will fuel a real revival of literary journals and print magazines full of thoughtful essays and serious book reviews.
At least it seems this author only wrote this book, and he seems like a legit person if you click on his name. For all I know Tom Lesley doesn't even exist.
Now anyone lazy enough (to even not remove the phrase "As an AI language model I can't..." from the generated text) can "write books". And these "books" will be fundamentally plagiarism and a bunch of trivia by design.
Not only have YouTubers found the same "business model", but now there are also people "educating" other people to "write" e-books and use image generators for book covers. Of course, you get also the AI generated promotion material with these courses.
The funny part is that he didn't even care to generate a different number of reviews for each book, or to add a few 4's in just to make it more realistic. He used the exact same number of reviews for many books, and all are full 5's!
That would be the result if it was automated, such as either a script or the AI itself instructed to write books on various subjects, selling them on Amazon, and then reviewing them. I don't see what can be done to stop this trend since it's 100% certain it will become ubiquitous.
So many questions: Are these books copyright by Tom Lesley, ChatGPT, OpenAI, ...? Should he share the money with OpenAI, or with the internet? Do we need to update the definitions of plagiarism? So it goes.
This [1] is, to my knowledge, the latest word on who'd own the copyright. And the answer is most likely the author, with the second most likely candidate being nobody. In a nutshell, "AI" generated content itself cannot be copyrighted since the process is carried out by software instead of a human, and software cannot be assigned ownership of copyright. However, material that is sufficiently recombined in a 'creative way' can be copyrighted.
The case/example mentioned is that a single image from image generation software cannot be copyrighted, but a series of related images being used to tell a story can be copyrighted, because it's a human carrying out the creative task.
> software cannot be assigned ownership of copyright.
More precisely, only human works are subject to copyright. The law simply doesn’t apply to works by animals, AIs, or anything else, even when they are as creative as a human.
So the next business model is to simply copy every work that isn't copyrightable and put it up for sale yourself. Nothing stopping you and you may as well.
Yes, but copyright requires the human to have significant original input into the work for it to be eligible. The human can’t just delegate to the animal or AI.
The degree to which a human has to be creative is very minimal. As SCOTUS said in Feist v. Rural:
> "the requisite level of creativity is extremely low; even a slight amount will suffice"
Literally throwing paint at a canvas is enough to suffice.
Arranging the prerequisite conditions and coaxing a monkey into taking a selfie is probably enough as well (as long as you don't hire PETA's dumb lawyers who want to argue for assignment of the copyright to the monkey)
There's not really enough information to determine whether this author would qualify. We don't know how much editing he did if any, nor have we seen these specific cases tested in a court yet.
I think it needs to be evaluated relative to the volume of the work. If you take a novel in the public domain and just change a few sentences, I don’t think you can claim copyright for the modified novel.
The courts are going to always be obsessed with one practical scenario. I take e.g. Harry Potter, put it through an LLM with the prompting of essentially writing the exact same thing with slightly different characters, locations, and phrasings. And then put it up for sale (or free download), making no secret of the fact that it's essentially the same book.
If an LLM reaches a sufficient capability that this can be achieved while maintaining a roughly comparable level of readability, then you'd have just completely legalized 'piracy by proxy'. And the same would come to every single medium from movies to software. This is going to make it an extremely difficult question to answer.
I expect what it come down to is determinism. LLMs are completely deterministic - same model + same input + same seed = same output. And so any seller will be obligated to retain any model composition/data, and any copyright holder of an input training item will have a copyright claim on any output item. In other words, you can only train on stuff you already have the copyright to, or rights to.
There maybe some need to legally clarify how much input is necessary for Tom to claim copyright.
ChatGPT can't, legally speaking only humans are capable of claiming copyright. In the monkey selfie case a few years ago it was determined animals and machines can't claim copyright. In cases where there is no human author, the work is public domain.
Isn't there a difference between a monkey taking a picture (no human author involved) and a person using a tool (ChatGPT) to help them realise their original work?
If I made a drawing with one of those old plastic spirograph tools I'd never wonder if the copyright was mine or the spirograph's.
Sure, even though photos are machine made, Humans claim copyright because they had artistic input on composition and framing. Between fully AI generated stuff, and human artistic input prompts causing output there is a line where the results go from copyrightable to not copyrightable. The question is where is that line, legally.
To me GPT is either very close to or already crosses that line.
OpenAI (ChatGPT, DALL-E, etc.) explicitly assigns their rights to anything generated by their systems to the user.
> OpenAI hereby assigns to you all its right, title and interest in and to Output. This means you can use Content for any purpose, including commercial purposes such as sale or publication, if you comply with these Terms.
So obviously what he's doing is unethical. The quality of the information is pretty bad...
But what about, say, selling a coffee table art book. "Visions of the Future" - like 50 good quality scifi images with captions generated by, oh, Midjourney?
15-20% profit margin on a $20 print on demand book... Wouldn't do it myself but it doesn't feel wrong? I assume people would pay for that just for the art itself in a physical form, it doesn't have accuracy requirements or anything..
Nonsense book scams is a time-tested Amazon tradition. I just hope that the gibberish won't contaminate the free/pirate archives (which are much harder to profit from).
Hm... The back cover of the data mesh book reads: "Data Mesh: Architecting Data Infrastructure for Agility and Innovation with Data Mesh" is a groundbreaking book that presents a new paradigm for managing data at scale. Written by Zhamak Dehghani, a software engineer and thought leader at ThoughtWorks, the book ..."
Amazon then recommended me the almost-as-prolific Caroline Manta. She's either a real person, or the scammer went to trouble of putting up a LinkedIn profile saying she's looking for web design work. I feel sorry for anyone who didn't know better handing over real money for this trash.
“If only there was a way to somehow detect all this spam and garbage on out platform” - Amazon’s tenured principal engineers with 500k salaries, probably.
“Try banning items where every review is 5 stars and every reviewer has a middle name for some reason.” - me to the rescue.
It would create * an inflation in the offer; * a staggering amount of noise; * a bubble that would conclude with the creation of services like "Instant book generator".
(I am not even mentioning the next step, to avoid seeding bad ideas.)
I’m not a lawyer, but I really doubt it’d be illegal in the US at least.
Just creating a book from ChatGPT is not against its terms of service, as “ OpenAI hereby assigns to you all its right, title and interest in and to Output. This means you can use Content for any purpose, including commercial purposes such as sale or publication, if you comply with these Terms.” What would be against the TOS is claiming that the output is human generated, so theoretically OpenAI could sue.
The courts have recently held that AI generated images are not copyrightable, so these “books” are basically just public domain content (where they aren’t directly copying existing human generated content). It’s not illegal to republish public domain content as your own.
The thing that might be straight up illegal here is the reviews. Those appear to be completely fake and probably constitute fraud.
> OpenAI hereby assigns to you all its right, title and interest in and to Output
How can it do that? ChatGPT isn't capable of owning or assigning copyright, because only humans can produce copyrighted works (or enter into copyright assignment contracts - typical in employment contracts), and ChatGPT isn't human.
That's a standard phrase to make lawyers happy. It means what whatever else happens and whoever else may claim title to the work OpenAI won't come after you. They are free to put this in their terms of service; many companies will refuse to use it without such a disclaimer.
Where it gets entertaining is in Section 7. Indemnification; Disclaimer of Warranties; Limitations on Liability, which contains the following gem:
(a) Indemnity. You will defend, indemnify, and hold harmless us, our
affiliates, and our personnel, from and against any claims, losses, and
expenses (including attorneys’ fees) arising from or relating to your use
of the Services, including your Content, products or services you develop
or offer in connection with the Services, and your breach of these Terms
or violation of applicable law. [0]
Unlike normal Terms of Service where the vendor accepts liability and/or indemnifies the user against IP violations or other issues, this clause turns things around. If there's a problem you the user must mount a legal defense for OpenAI. By using Chat GPT you are enrolling in the legion of its defenders.
And obviously he did not really even attempt to hide anything. Whether it's legal or not: I don't think we have a clear, 100% obvious understanding of the consequences right now, but I'm pretty sure OpenAI couldn't care less. Suing people who use ChatGPT to publish meaningless books would surely not do them any good. Who else might care enough?
The Copyright Office has decreed that AI generated content cannot be Copyrighted. If these books have a copyright listing then they are a violation of the rules. I doubt if the "author" is at risk for legal repurcussions, beyond not being able to sue for copyright violation.
There is nuance in the statement that you're referring to. Works created by an AI are not copyrightable, but works created by humans who use AI tools are copyrightable.
Non-humans do not have copy rights from the things they create. However, humans have copy rights even when they use tools.
I have yet to see a legal argument that says the output of ChatGPT is free from derivative works claims. So you can't assert ownership of the output, but that doesn't mean someone else can't assert rights on the grounds of derivative works.
It is one of the reasons I don't see Copilot going anywhere. Any company that uses it can't assert ownership over their own IP, and if Copilot accidently reproduces a chunk of training data verbatim then someone else can.
I doubt many codebases are 100% generated by Copilot with no human input, and it’s not like having some Copilot snippets in your code is going to invalidate the copyright on the whole thing… I think that’s like being worried that writing parts of a book in a public domain font is going to invalidate the copyright on it
The copying verbatim thing is valid though, although they have mitigations against that
From one review: “ One paragraph starts with "As an AI language model I can't....". Unfortunately we don't have an AI who can correct a book written by an AI. Fortunately I can send it back.”
Sauce:
https://www.amazon.com/product-reviews/B0BWVH451G/ref=acr_dp...