The world’s top two AI startups are ignoring requests by media publishers to stop scraping their web content for free model training data, Business Insider has learned.
OpenAI and Anthropic have been found to be either ignoring or circumventing an established web rule, called robots.txt, that prevents automated scraping of websites.
TollBit, a startup aiming to broker paid licensing deals between publishers and AI companies, found several AI companies are acting in this way and informed certain large publishers in a Friday letter, which was reported earlier by Reuters. The letter did not include the names of any of the AI companies accused of skirting the rule.
OpenAI and Anthropic have stated publicly that they respect robots.txt and blocks to their specific web crawlers, GPTBot and ClaudeBot.
However, according to TollBit’s findings, such blocks are not being respected, as claimed. AI companies, including OpenAI and Anthropic, are simply choosing to “bypass” robots.txt in order to retrieve or scrape all of the content from a given website or page.
A spokeswoman for OpenAI declined to comment beyond pointing BI to a corporate blogpost from May, in which the company says it takes web crawler permissions “into account each time we train a new model.” A spokesperson for Anthropic did not respond to emails seeking comment.
Robots.txt is a single bit of code that’s been used since the late 1990s as a way for websites to tell bot crawlers they don’t want their data scraped and collected. It was widely accepted as one of the unofficial rules supporting the web.
Oh boy, if they’re ignoring robots txt, then I better …add a useless link at the bottom of every comment I make. That’ll really show them!
This comment is copyrighted by me and licensed to the public under the terms of the CC-BY-NC-SA 4.0. If you intend to use this comment for commercial purposes, you must secure a commercial license from me, which will cost you a lot of money. If you violate the terms of the CC-BY-NC-SA 4.0 without securing an appropriate license, I will send my army of lawyers that I totally definitely have to defend my copyright against you in court.
Exactly, I never understood what people thought they would achieve by putting the link to that in their comments. Like, AI firms are absolutely willing to skim through copyrighted works of artists, backed by a much stronger license, what makes you think linking that will achieve anything. Except maybe poisoning the LLM well.
Hey, there’s a thought. If we all just put that at the end of every comment, I wonder if GPT6 will figure that’s just how people talk and end all it’s responses with it?All they’re going to do is teach the AI that sometimes people end posts with useless disclaimers.
I’ve been buffalo buffalo buffalo buffalo buffalo trying to remember to put some nonsense somewhere in my comments every time in order to make the LLMs think this is how people talk.
I’m anaspeptic, phrasmotic, even compunctuous to have read such pericombobulation.
That’s quite quiet quintessentially quiescent and clearly colloquially colour clever.
They figure no one person will have the means to sue or the ability to prove that their data was scraped.
Could we take em to small claims for like $500 a comment? That would be a devastating movement lol
This media provided under the CC-Mine-Now license. You may remix this comment for use in your AI feeding but if you do I own your company now and all proceeds go to me, and you indemnify me against anything I want to do to your company.
The real problem is robots.txt is an honour system in the first place - It’s never been a defence against bad (or even simply poor faith) actors.
The game plan is to scrape, store and utilise as much data as possible regardless of conventions, best practice, license agreements etc until specifically regulated to stop.
At that point, a few early companies will have used vast swathes of data that any newly established company is banned from also using
And they will be “unable” to purge it.
Hoping the EU drops GDPR 2 requiring them to delete the entire model if it infringes or something.
Expecting the US to meaningfully regulate US companies is like expecting…
You know what, even including physical impossibilities, I’m struggling to think of anything less likely
I’ve yet to understand how the hell they get away with “I don’t know how it works”. Either figure out how it works or stop using it, shithead. It’s software not magic beans.
There’s lots of complicated fields out there, none of them get a pass for “I don’t know how my drugs work” or “I don’t know how my rockets work”. That’s absolutely ridiculous.
Uh, we don’t really know how our drugs work (especially the older ones). We have a vague understanding of their mechanisms, but we really don’t know how they work. We don’t even have a clear idea of what the structures of most drugs look like, and how they interact with their binding sites.
Luckily, we don’t actually have to know how they work, to know that they work. Instead we use clinical trials and real world evidence to support their use.
(Fun fact: there’s actually a branch of drug development called phenotypic drug discovery which actually does away with the understanding of the mechanisms altogether. )
It’s just how machine learning has been since ever.
We only know the model’s behavior by testing, hence we only know more or less the behavior in relation to the amount of testing that was done. But the model internals has always been a black box of numbers that individually mean nothing and if tracked which neurons fire here and there it’ll appear just random, because it probably is.
Remember the machine learning models aren’t carefully designed, they’re just brute-force trained for a long time and have the numbers adjusted again and again whenever the results look closer or further away from the desired output.
If the models are random then we shouldn’t be trusting them to do anything, let alone serious applications. If any other type of software told us that it’s based on partially random results we’d say “get that shit out of here, I want my software to work first time, every time”.
“Statistically good enough” works for some applications but not for others. If a LLM finds a formula that has an 80% chance to be the cure for cancer or a new magical fuel or some amazing new material that’s cool, we’re not going to look the gift horse in the mouth.
But using LLM to polute the web with advertising texts that are barely inteligible, and using it as a pretext to break copyright in the process, who does that help? So far the only readily available commercial application for LLMs has been to spit out semi-nonsense so that a bunch of bottom-crawling parasitic industries can be enabled to keep on pinching pennies and shitting up everything they touch.
Which, ironically, it will help them to hit bottom all the faster, so in a strange way it’s a positive return, but the problem is they’re going to take down a lot of useful things with them.
If the models are random then we shouldn’t be trusting them to do anything, let alone serious applications.
That’s not the reason we shouldn’t be using them for anything other than generating lorem ipsum style text or dialogue for non quest critical NPCs in games.
The reason is that, paraphrasing Neil Gaiman, LLMs don’t generate information, they generate information shaped sentences.
Specifically, an LLM takes a sequence of characters (not a word or text; LLMs have no concept of words, or text, or anything else for that matter; they’re just an application of statistics on large volumes of sequences of characters; no meaning or intelligence involved, artificial or not)… as I was saying, an LLM takes a sequence of characters, pushes it through its model, and outputs the sequence of characters most likely to follow it in the texts its model has been trained on (or rather, the most likely after discarding the ones its creators have labelled as politically incorrect).
That’s all they do, and they’ll excellent at it (or would be if it weren’t for the aforementioned filters), but that’ll never give you a cure for cancer unless there already was one in their training data.
They take texts written by humans, shred them, and give you their badly put back together dessicated corpses, drained of any and all meaning or information, but looking very convincingly (until you fact check them) like actually meaningful or informative texts.
That is what makes them dangerous. That and the fact that the bastards selling them are marketing them for the jobs they’re least capable of doing, that is, providing reliable information.
(And that’s while they can still be trained on meaningful and informative texts written by humans — inasmuch as anything found on reddit, facebook, or xitter can be considered to be meaningful or informative —, but given that a higher and higher percentage of the text on the internet is being generated by LLMs soon enough it’ll be impossible to train new models on anything but 99% LLM generated garbage, at which point the whole bubble will implode, as anyone who’s wasted time, paper, and toner playing with a photocopier or anyone familiar with the phrase “garbage in, garbage out” will already have realised… which is probably why the LLM peddlers are ignoring robots.txt and copyright laws in a desperate effort to scrape whatever’s left of the bottom of the barrel.)
LLMs don’t generate information, they generate information shaped sentences
That is besides the point. A random number generator is more or less random but it still has applications.
The problem is not them being random, it’s hiding that they are being random so they can be used for applications where randomness is not a feature.
I’m in the US so yeah…. Even if the current of future GDPR requires deletion I guarantee it’ll still be used in the US. I have no faith that any US company will follow rules like that. Any fines are just looked at as the cost of doing business.
Or they’ll “purge” it and somehow the canaries will end up in the model anyway
It’s like weapons testing. You only move to ban testing after you’ve developed it yourself.
Whatever happened to those “nightshade” images that poison the model?
You mean that work that took open source software, closed sourced it and refused to release the source code and the poisoning only worked against one specific open source model (stable diffusion)? I don’t think that’s going to come riding to anyone’s rescue.
They only kinda work but more importantly they need mass adoption to actually poison training data. Most people aren’t going to add another step to their posts so probably the only way to mass adopt it is to have platforms automatically poison uploaded images. I wonder if reposts on a platform like that would start to have noticable artifacts in the images like jpeg but different
Same approach as all the other ‘disruptive’ new companies that ignore industry standards, rules, and laws.
I’d say they are pushing for regulations behind the scene because they know it gives them an instant monopoly.
They are already pass the door, they can afford to shut it behind them to own the room. Having to send checks to websites like Reddit and Getty in the future is a small price to pay.
Am I missing something in this article? I’m not defending either company, but it doesn’t seem like they actually have any evidence to confirm either is doing this.
The world’s top two AI startups are ignoring requests by media publishers to stop scraping their web content for free model training data, Business Insider has learned.
It claims this, but then they say this about the source of this info:
TollBit, a startup aiming to broker paid licensing deals between publishers and AI companies, found several AI companies are acting in this way and informed certain large publishers in a Friday letter, which was reported earlier by Reuters. The letter did not include the names of any of the AI companies accused of skirting the rule.
So their source doesn’t actually say which companies are doing this, but then they jump straight into this:
AI companies, including OpenAI and Anthropic, are simply choosing to “bypass” robots.txt in order to retrieve or scrape all of the content from a given website or page.
So they’re just concluding that based on nothing and reporting it as fact?
So cynical … what makes you think “a startup aiming to broker paid licensing deals between publishers and AI companies” can’t be trusted implicitly?
A spokeswoman for OpenAI declined to comment beyond pointing BI to a corporate blogpost from May, in which the company says it takes web crawler permissions “into account each time we train a new model.
The translation for this is do we stand to profit more than we stand to be punished.
Basic capitalist risk assessment in other words.
They can’t even be punished.
robots.txtis just a convention, not a regulation. It’s totally not enforceable.The only legal framework we have is copyright law. Those who oppose this behavior will have to demonstrate copyright violation, and that may be difficult to do since the law hasn’t caught up.
It’s true robots is not regulation but if it’s proven they ignore it on purpose it will be a major point in future lawsuits. And those are the next step.
It won’t have any relevance at all.
Either scraping to transform the information in the page is fair use, and consent isn’t necessary, or it is not fair use, and the absence of a robots.txt doesn’t constitute consent. There’s no middle ground where a robots.txt can mean anything.
Yeah I know. But I wanted to point out that the comment in the article wasn’t so much a real consideration as business risk analysis 101. Along with a healthy dose of corporate spin.
Robots.txt isn’t even a rule, it’s a request.
“Please do not ask for the following content if you are a robot”.
If you don’t want someone to look at your content, you ultimately have to not give it to them, not just ask them to not ask.
They stand to profit if this is made into a real law.
Any regulation on AI just kill off their competition at this point. They are both lobbying for it and numerous proposed “anti-AI” laws have been their doing.
TollBit, a startup aiming to broker paid licensing deals between publishers and AI companies.
If we can’t scrape data freely, it instantly kills the open source scene. These regulation only benefit companies like OpenAi and Google, who will happily pay an exorbitant price to have exclusive rights on data they don’t already own and get a monopoly in return, as well as the companies who own this data like Reddit, Getty, Adobe, etc.
Getting a dime was never in the cards for individuals except maybe the outliers like GRR who can throw their weight around.
Almost all regulation being proposed only benefit big AI companies and are meant to kill any competition. They are flooding the media with bad sentiment articles to manipulate people so they can tell congress their constituents want this.
Exactly.
If you can’t train using public, copyrighted material, Disney has a hell of a model and their monopoly over the entertainment industry goes from huge to insurmountable. No “little guys” gain anything. It’s regulatory capture, nothing more.
Voice of reason
So, the same thing media websites do when they ignore my “do not track” request?
Generative AIs are thieves and this is just more evidence.
If you put it online, you gave it away. If someone reads it and then uses that information to answer questions for someone else without giving credit to the author, that’s called a conversation. As long as no copyrights are being abused, there is no problem and this is just corporations upset with what they think is piracy, pandering to people who are still on the fence about AI.
Except we shouldn’t be giving corporations same rights as individuals. Doing so leads to corporate feudalism.
So it’s okay for me to pirate something but not a corporation?
OpenAI has a clause that one cannot train their own AI on OpenAI chatbots
If it was all a giant open source project I’m sure many would be more accommodating to your argument.
So if it’s open source, it’s okay?
You got it backwards. According to OpenAI and Microsoft you have to respect their copyright but they can ignore yours.
Also no you can’t pirate but they can.
Any questions?
Id say yes to that statement, but for reasons that dont have to do with AI as I dont really view AI training as piracy.
Novice web site owner/coder here: wondering if I can block them somehow via IP address in addition to robots.txt. Server firewall rule? Remember, I said I was a novice…
You can block an IP but first you would need to know which IPs are scrapers. And they could just use a VPN to bypass IP blocks.
deleted by creator
Yes, the less expensive VPNs especially have a lot of users using the same IP addresses.
You can get a VPN with private IP’s but this is more expensive. For a company of OpenAI’s size that would be a drop in the bucket though.
Google “spider trap website” or something.
Yes, they’re evil. Was there a question?
