I understand it prevents publishers' works from being "stolen" but leaves them without a major distribution source.
Even if there are people, today, that do publish valuable stuff on the internet only because (or thanks to) the expected traffic (either own ads or selling ad space), I claim that without that traffic, there will be others who will be publishing valuable stuff, and the traffic incentivized garbage more than the valuable stuff.
Finally, if that's not the case, then in order for the LLM to be useful, it needs valuable training data. So this problem solves itself - competitive LLM company will need to pay for the data directly, and for me, that model incentivizes valuable stuff more than garbage.
In the meantime, I think people should focus on attribution, and algorithms to find related work (which may suitably substitute for the former). This will allow us to fund creators and publishers for AI output, maybe by forcing the AI companies, or naturally through patronage (see AI output you like, find who owns the training data that contributed to it, donate to them). Moreover, it will help people discover more interesting things and creators.
I think if you wanna break up Google it would be a lot more effective to separate search from AdSense, rather than chrome.
Obviously there are massive profits to be made from private browsers.
Is going to print money like google ads? Probably not. Could it sustain the browser as a standalone business? Perhaps.
Being a monopoly is not unlawful.
"... the possession of monopoly power will not be found unlawful unless it is accompanied by an element of anticompetitive conduct."
https://www.courtlistener.com/opinion/131153/verizon-communi...
Google's anticompetitive conduct occurred mainly through agreements concerning web browser defaults.
Controlling default settings in web browsers via these agreements will be prohibited. It stands to reason that controlling default settings by distributing its own web browser will also be restricted.
"Maybe off topic, but Ive always found strange people who comment on Google's monopoly by focusing on chrome."
As such, it is not strange that proposed remedies targeting Chrome were submitted and have been a focus of online commentary.
Fix the insanity that is intellectual property and get back to us.
Was it the responsibility of the people who created automated, cheaper and better elevators to defend the job of these obsolete people ?
Not really.
If they would have respected the laws, OpenAI would not have been able to create AI models at all (as they are derivatives of copyrighted works).
So it's pure piracy of content at scale. The same with Veo3.
Is it beneficial to humanity ? Probably yes.
Is it harmful to content creators ? Absolutely but in the long-term, they have little chances to survive, no matter if legal or not, because this is both what the market and consumers want.
For example, these people on Fiverr selling blog posts, it was minimum 200 USD per blog post, and few days of turnaround.
Now with AI it's 0.01 USD and instant.
I think in the long term the highest quality content creators are going to find ways to keep their information out of AI training data, and put it behind walled gardens.
The problem I see is that when everything is iterative, and you kill sustainable incentives for new views, innovation disappears.
Is like the movies. We will get all budget towards remakes and reboots, and little new would show up.
Being an 80s guy, I remember many new movies coming out every year, and when they stretched beyond 2 or 3 films, people frowned.
Back to the future, Indy (the good years), Rambo (first 3), lethal weapon. Cool gems came up too (Last action hero?Terminator) even in loaded fields like simple action movies…. Star Wars, 3 films. Matrix. bond was the exception.
Now? mission impossible (8 films at over 2h), fast and mediocre franchise (20 movies), Marvel and DC movies (3 movies shot 30 different times), and everything else is a reboot. Almost impossible for new ideas to get through.
We are killing culture, innovation and human spark for some companies to make a billion more. We are feeding like factory pigs from generic slop, and as the factories grow they kill fertile ground for those who wanted to eat something cleaner.
The fact that is happening does not mean it’s right. It’s just we are accepting mediocrity as our guiding principle at 0.01 USD a pop…
The content and services provided by AI still needs those humans behind. Maybe when it’s all AI agents talking to themselves, this would make sense, but before that, someone has to write, edit and choose that content.
How can you say publishers are obsolete ?
And no one expects a "free job"; they just expect their work to not be used in ways that doesn't benefit them at all (but instead benefits someone else).
The bizarre analogies AI maximalists are pushing are boring and tedious to the point of bad faith.
But what do I know ... I'm just an "obsolete person" shrug
We’re talking Future Publishing or even NYT here…
11.6 million subscribers. 600,000 of them get the dead tree edition.
https://www.nytimes.com/2025/05/07/business/media/new-york-t...
Because of the variance in production cost and ROI for effort, vs the general ability to copy IP (just the IP, which could be loads of research that isn’t valuable without loads of capital and decades of investment into certain infrastructure e.g. TSMC), there have to be some sort of incentives to do the hard groundwork that creates valuable IP that once discovered is trivial to copy or implement, otherwise people simply won’t do it, they’ll just copy and resell and then fewer people will do the initial leg work until we’re left chasing the bottom of the barrel of the level of effort someone may invest in something they have no means of capturing return from that work and risk. And not everything can be hidden behind a web service, some IP is far more challenging to protect.
So, while I tend to agree that the current implementation of IP protections is trash, I think we do need something there to encourage such work. In research we have(/had?) federal tax investment into doing such work and as a trade off, researchers get fed and live a semi comfortable or sometime decent quality of life for their efforts. I don’t have the answers to the ideal model, I know what we have isn’t great and can be improved upon in some ways, but I don’t know wha the best(TM) solution is here. I know it’s not pure unrestricted markets doing whatever they want, though.
It's not a stupid concept. It exists because someone put effort in. Should we steal their effort?
Imagine if Wikipedia had a paywall for each article… but you had to pay to a different organisation for each vaguely related set of pages. Some requiring recurring subscriptions and minimum spend so you can’t cancel.
Worse, the people demanding the payment aren’t even the operators of Wikipedia and they charge the article editors for the privilege of publishing on their platform!
You can have government funded research at a government funded university and then both the researcher and the students at that same university have to pay to have the research published and accessed!
But not too many pictures please and they can’t have color, that’s expensive, you see?
Raw data hosting? No way! Do that yourself.
Versioning, backlinks, ongoing feedback, comments, etc… don’t even ask! Do you think this is the future!?
Meanwhile the largest collaboration of human minds on the planet takes place on GitHub and is almost entirely volunteers working for free and giving away their work for free for everyone to fork and build on. And comment, contribute, and extend.
Science as it is today is a tyre fire and it is almost entirely the fault of rent seeking middlemen with misaligned incentives.
Somebody does something which takes some work. There's nothing special about his work other than that it is available-- there's so many stories you can tell, there's so many ways to express an idea, that to, if the author hadn't created a particular work, making it illegal to create that particular work, people wouldn't have lost anything. It would be a trivial restriction.
Thus when someone takes a particular work and does something with it, he chooses to not develop one of the other similar works he himself could have created, and takes advantage of somebody else's effort.
Western governments would rather publishers lose some property rights than tolerate a world where their industries become dependent on Chinese LLMs. Although to be clear, they probably don’t enjoy either option.
Western governments would rather citizens lose some rights and years off their life than tolerate a world where their industries become dependent on Chinese manufacturing.
————
Western governments could make models trained on stolen IP toxic if they wanted.
It may not be worth it. The trade off of model capability vs the value of IP law might be worthwhile.
But western governments should show a little dignity and openly make that trade off. We don’t need to appeal to “but China”.
It’s not clear that was a good decision, but tariffs and similar restrictions at least provide hope of a remedy. But we can’t use tariffs on open source software.
Their motivation wasn't the maintenance of better environmental and labor laws, but higher profits from exploiting China's more business friendly (less regulation, fewer labor rights) environment. Had it been economically and politically viable for Western governments to domestically enact the same environmental and labor standards as China, they would have done so. And as soon as the quality of life in China improved enough that labor there was no longer economically viable, many companies moved their outsourcing elsewhere, such as to Vietnam or Africa.
Tl;dr* He claims ads were the original sin of the web, built for a human internet. For the coming agentic web where most browsing would be done by agents, he proposes a new protocol that has payments integrated into it. Specifically agents paying for accessing content using micropayments with stablecoins.
Also interesting to note that this is one of the most submitted articles of Stratechery to HN but gained no traction. Latest submission: https://news.ycombinator.com/item?id=44073241
* A summary can also be found here: https://stratechery.com/2025/the-upheaval-coming-for-the-int...
Tying it to cryptocurrency just seems like it's even less likely to get your regular Joe to onboard to this scheme, so I can see why it's never gained wider interest.
Honestly, some form of digital credits work just as well here IMO (barring regulatory hurdles). No crypto necessary.
ivape•21h ago
jxjnskkzxxhx•20h ago
I dont think that's how LLMs work.
enjoylife•20h ago
Cyclone_•20h ago
jxjnskkzxxhx•18h ago
ivape•17h ago
jxjnskkzxxhx•9h ago
Please explain to me how an algorithm that takes a sequence of tokens and outputs a URL would work.