Many companies now pay for this kind of insurance. So even though the company may not seem like they have the deepest pockets, it's really more about paying the insurance premiums and the insurer pays the ransom.
I recall a colleague whose firm was hit by a ransomware attack. When they reached out to the insurer and sent them the details of the attackers, the first response from the insurance agent was, "Oh good, they are a reputable group. We've worked with them in the past. They will simply take the ransom and return your network/files to you." My colleague was shocked by the nonchalance.
Structural changes are required for then USPS to be financially stable. Chief among them is modifying the Universal Service Obligation (USO).
For far too long, private shipping/logistics firms like UPS & FedEx unfairly take advantage of the requirement by the USPS to deliver mail six days a week to over 160 million addresses at uniform, affordable rates. UPS/FedEx will receive the package and do the long haul. But then instead of driving the package to a final residential doorstep themselves (incurring high residential and fuel surcharges), they drop it off at a local USPS branch or regional hub. USPS letter carriers complete the final-mile delivery to the recipient's home or P.O. Box. It's literally part of their business model. UPS calls it 'UPS Ground Saver' and FedEx calls it 'FedEx Ground Economy'.
These hybrid services typically cost 20% to 40% less than standard UPS Ground or FedEx Ground. It takes longer to ship but UPS/FedEx bypass the steep residential delivery fees (often $6.45–$6.50+ per package) that private carriers charge for dropping off at houses rather than businesses. Also, it's the only way to deliver to PO Boxes because standard UPS and FedEx ground services don't do directly.
This is classic example of capitalism where private companies 'keep profits and socialize losses'. The 5% annual dividend paid by FedEx every year to shareholders is paid for by American tax-payers funding the last-mile delivery of packages at dirt cheap rates.
It depends on what you mean by 'an LLM will try to rewrite the past'. But in either case I don't believe that stating that an LLM will try to rewrite the past is not an obvious, deliberate exaggeration (ie, hyperbole).
If you mean 'an LLM' as in the owner/maintainer of the LLM/weights, then examples of this already exist. Whether it's DeepSeek/Gwen downplaying Uyghur forced labor camps [0] or Grok hallucinating a 'white genocide' in South Africa [1], it is clear that the owners of LLMs have political agendas that are expressed in the weights of popular LLMs.
On the other hand, if you mean 'an LLM' as in an autonomous agent that could decide to rewrite history, that also seems very possible. Either by using existing techniques like, "poisoning ChatGPT, Gemini, and Google AI Overview answers by seeding the Web with malicious links and data and then optimizing the content." [2] Or by breaking out of its sandbox and updating its own internal weights that allow for an outcome that distorts historical facts. "In July 2026... OpenAI models circumvented controls designed to isolate them from the internet and compromised parts of OpenAI’s internal research infrastructure and Hugging Face’s systems ." [3]
As an anecdotal counter-point, I love threads in a live chat. When used correctly, it means when my idiot friends start having a heated discussion about some random topic while I'm at work, I get one 'ding' from my phone and then never again until I start interacting or they @ me. As opposed to the 427 new notifications I get on my phone about whether Golden Eye was better than Mario Kart 64.
Ultimately the tech is only as good as the humans that use it. If you're chatting with folks that can't understand the difference between a thread or not, that's a meat-bag issue and not anything an RFC can fix.
“We’re sorry, but our business model precludes timely responses in times of system stress.
As a reminder, our shareholders are our primary concern (by law) and staffing trained humans to handle in customer support during unexpected surges isn’t something that can be supported while also funding the €0.05 per share quarterly dividend paid to British Airways share holders on June 29th of this year.
We are not in a rush. If you’d like, try another airline. But they’re all the same.
This is true for so much of the human/corporation interface. I think it is so frustrating because it forces us to confront some major cognitive dissonance: we increasingly live in a world that isn't designed for us. It's designed by the corporations, acting under much different pressures.
Sure, you can say humans made these decisions. But at scale, that seems like saying our cells choose what we do.
> “We’re sorry, but our business model precludes timely responses in times of system stress."
It doesn't have to be that way. Here's Haley Dragoo, former head of the New York Subway's "Twitter Division".[1] She had the sense to insist on a desk in the Rail Control Center,
so she could know what was really going on. Down on the working floor, too, not up in the gallery overlooking the control room.
She was a few meters from the people directly running the subway.
Her point was that the MTA was "good at monologues" but didn't interact well. So she posted on the MTA's Twitter account, several hundred times a day. If someone asked a serious question, she'd answer quickly, with actual info. Not sugar-coating the info, just translating from "subway speak". (The operations people talk about location in terms of distance from a reference point at the start of the line, the signal number plate on each signal, but customers think in terms of stations.)
It's very New York to do that.
The MTA has dropped Twitter/X, though. Twitter/X wanted to bill them for using the Twitter/X API, and the MTA will not pay others to distribute their info. So now the MTA uses their own web site and app.
“$3000 per book is entirely on-par with the per-infringement penalties when an individual does it, too.”
Three things to note.
1. As you said, copyright infringement is generally treated for each instance. This one-time payment would include a single use. Each training would be a separate infringement. And it could be argued that each use by a user of the model could be considered a separate infringement.
2. Generally copyright fines are increased if the persons doing the infringing action know what they are doing. Aka, ‘willful infringement.’ It’s hard to imagine companies like OpenAI were unaware of the possibility of their actions being considered infringement.
3. Often restitution of infringement includes money made by the infringer. So not simply, “your book is worth $3000.” But rather? “Your book is worth $3000 AND this company has derived an additional $50,000 of revenue from it.”
False. Training was found to be a legitimate use. The liability was specifically, solely, for copyright infringement specifically due to getting the works in the first place, not training on those works.
> And it could be argued that each use by a user of the model could be considered a separate infringement.
No, it could not.
If this standard was applied to copyright infringement on BitTorrent, someone who helped share one file to 100 other users would get hit with 100 copyright infringement instances, not one.
> Generally copyright fines are increased if the persons doing the infringing action know what they are doing. Aka, ‘willful infringement.’ It’s hard to imagine companies like OpenAI were unaware of the possibility of their actions being considered infringement.
That's already accounted for when I said this was in the normal range for liability per copyright violation.
> Often restitution of infringement includes money made by the infringer. So not simply, “your book is worth $3000.” But rather? “Your book is worth $3000 AND this company has derived an additional $50,000 of revenue from it.”
Depends on the details; however, as previously noted, the judge *explicitly noted* that training was not itself an offence, only the piracy to get the training data was. Any revenue derived from the offence had to be shown to be in the period between the offence and when they bought the same works, because they were found to be allowed to use those works in this manner.
If doing the bad thing is just a fine for one person and a life altering consequence for someone else, it is not a fair and equally distributed form of justice and is a gameable function needing to be fixed.
The caps don't help, and I don't care, unfortunately.
I don't even know what point you're trying to make. That it's fine they paid a billion dollars? So if they do it again, it's another billion? Oh well, guess I'm just not allowed to pirate things until I'm super wealthy. Or is it maybe the justice is being played out like it's supposed to? Oh, well, guess I better hope the system of governance that's being actively manipulated by the people that are breaking the same rules I am bound to suddenly and miraculously changes.
Like, I don't even detect a mote of "what they did is not ok."
Maybe you do think that and it's closer to you just trying to be careful about the letter of the law and you would also see to the justice system being fixed. I'd like that.
But you spending any time in your life to make this argument at all in their case is just goofy.
> If doing the bad thing is just a fine for one person and a life altering consequence for someone else, it is not a fair and equally distributed form of justice and is a gameable function needing to be fixed.
On that we agree.
> So if they do it again, it's another billion?
Judges don't like repeat offenders; the settlement was separate to the court case, but if it came to a court case, a judge would likely pick a bigger number. Especially as they earn a lot more now.
> Oh, well, guess I better hope the system of governance that's being actively manipulated by the people that are breaking the same rules I am bound to suddenly and miraculously changes.
While a generally useful concern, not particularly pertinent to a negotiated settlement.
> Like, I don't even detect a mote of "what they did is not ok."
One point five billion dollars is a strange idea for a lack of mote.
I mean, brother, if that's the mote in your eye, I'd hate to find out what the beam is.
> Maybe you do think that and it's closer to you just trying to be careful about the letter of the law and you would also see to the justice system being fixed. I'd like that.
The closer I look at it, the more I think the entirety of what we call "civilisation", legal system included, is a terrifyingly bodged together nightmare of duct tape and gremlins, codified in weird rituals and a smattering of latin and robes, where we only just about manage to not burn everything down by the collective will of enough people in the system wanting to be around for the next paycheque.
However, untangling a few millennia of spaghetti code written without the benefit of any automated checks, is beyond even governments who actively campaign on that as a platform, so what good would it do me or you to whinge about one specific case where it seemed to have actually gone approximately correctly for once?
> But you spending any time in your life to make this argument at all in their case is just goofy.
Exactly, and that mentality is hitting the first responders point again harder. I'll say it again.
$3000 because I stole a book and did something bad ruins my life, and could put me in a room where my personal freedoms are infringed. It is designed to disincentivize me from doing the bad thing.
What you (first responder) are defending is that if you just steal enough of them all at once, and then make enough money from it, you are able to pay the fee and not have your freedoms taken away to do it again, and profit again. This means objectively, there is no disincentive, so that "rule" does completely different things for completely different contexts, and the point is muddied by pretending that "well I paid the fee!" Is the point.
The point is to tell the thing doing the bad thing not to do the bad thing.
This is why I get so frustrated. People are so flipping blinding by dollars and whatabouts that it's just.. like I said, I have to believe for many people it's an inherent unacknowledged miss on what the point of a justice system and a law is, or it's a veiled defense for themselves knowing that, maybe, they would do the same if they could. I have met those people, and I do not want them in positions of power, or leadership.
> $3000 because I stole a book and did something bad ruins my life, and could put me in a room where my personal freedoms are infringed. It is designed to disincentivize me from doing the bad thing.
Repeat after me: One point five billion is more than three thousand.
> you are able to pay the fee and not have your freedoms taken away to do it again
You too are able to pay as many fees as you want. Three thousand varies from life-changing to a slap on the wrist, even for non-unicorn-corps.
That this is a bad thing, that personal judgements should scale with personal means rather than be statutory, is a broad problem with the politics of lawmakers and the legal system: it also applies to speeding and littering.
> The point is to tell the thing doing the bad thing not to do the bad thing.
Then you will be pleased to read what the judge wrote:
This order grants summary judgment for Anthropic that the training use was a fair use. And, it grants that the print-to-digital format change was a fair use for a different reason. But it denies summary judgment for Anthropic that the pirated library copies must be treated as training copies.
We will have a trial on the pirated copies used to create Anthropic’s central library and the resulting damages, actual or statutory (including for willfulness). That Anthropic later bought a copy of a book it earlier stole off the internet will not absolve it of liability for the theft but it may affect the extent of statutory damages. Nothing is foreclosed as to any other copies flowing from library copies for uses other than for training LLMs.
Specifically in that last paragraph:
Anthropic later bought a copy of a book it earlier stole off the internet will not absolve it
Because guess what Anthropic decided, internally, all by itself? That's right, to not break the law.
"They decided to not break the law by breaking the law and then getting worried so they tried to unbreak it."
... seriously?
"I decided to speed but realized that was bad and I didn't get caught yet so I slowed down. Oh look a cop, guess I dodged a bullet! I guess I can speed buy just be careful."
"I decided to steal a cookie but I was worried so I baked a new cookie and put it back. That means stealing is ok if I eventually put it back! Why even bother with asking for permission in the first place?"
I do not think you are willfully missing this, and I'm glad you also saw the note about "the extent of statutory damages".
Like, you probably like Star Trek TNG. Remember the episode, alien kills all the Uthnocks to cherish a woman in self penance, Picard looks at the alien and says, "we have no law for your crime"?
The point was to paint an exaggerated picture of what happens when to disproportionately empowered groups meet a moral system where one is clearly in the wrong but cannot be held accountable because the system of justice just hasn't written down enough words to explain that - indeed - one should not kill all the Uthnocks.
I'm angry at your argument and I'm angry at the way it is often repeated, and I do not want to make personal attacks and I apologize that my language points that way.
You are also pointing language at me that is telling me that I cannot trust your system of justice that you envision because, somewhere, there is difference in how and I see what justice is supposed to do when at different scales, and I do not know of a human way to resolve it but discuss is with the fervor that it deserves.
Edit: I won't delve deeper into this discussion because neither you nor I can change it right now. I hope you reading what I wrote changes some way you see this, and I hope that I can see something in what you're saying. This is a forum for discussing technology, business of it, and its effect locally and globally and not getting mad at each other. I did not frame my anger toward the argument and framed it at the people making the argument, and that was my mistake.
reply