Feels in line for Anthropic who at least purport to try to be a more human-friendly company. Yeah, they did the thing everybody else did, but they're also leading the charge on repercussions which...
Feels in line for Anthropic who at least purport to try to be a more human-friendly company. Yeah, they did the thing everybody else did, but they're also leading the charge on repercussions which sorta works.
Hopefully this leads towards actually paying for the books even if it's one license being replicated by a single company for model training, as well. I'm curious what's going to happen with the existing models.
Enjoy the 30 grand! I don't know how much potential loss you incurred from having these works stolen but that's gotta be a pretty nice bit of news to get!
Enjoy the 30 grand! I don't know how much potential loss you incurred from having these works stolen but that's gotta be a pretty nice bit of news to get!
Was it published first in Europe or in the US? Are you a US citizen or no? There's some weirdness with copyright and international treaty. I can point you towards a Bluesky thread or two, but I...
Was it published first in Europe or in the US? Are you a US citizen or no?
There's some weirdness with copyright and international treaty. I can point you towards a Bluesky thread or two, but I think if it's published in Europe first, then no, you can't. But I'm not an international copyright lawyer
EWC @ Anthropic settlement proposal for U.S. copyright claims - EWC - European Writers Council What Authors Need to Know About the $1.5 Billion Anthropic Settlement - The Authors Guild A number of...
Registered within 3 months of its first publication (in any format) OR before the titles were downloaded (because the Copyright Act provides statutory (court-set) damages only if one of the two is true); AND
Registered within 5 years of first publication (because the Copyright Act provides a presumption of validity of all of the facts in the registration certificate, such as author and owner for works registered within 5 years).
A number of US authors are finding out that Publishers have had a very haphazard practice about actually registering the copyrights.
If your book is registered in the US it might be good. I am trying to track down the instructions. But there's a bunch of info in the links. I believe the "first publication" part is where it gets weird for the registration.
A good person to follow on Bluesky on this is Courtney Milan and the people she talks to - she straddles authorsky and lawsky and is a lawyer who clerked for SCOTUS and has spoken on how previous settlements have violated international copyright law so we're not great at this
This is a move in the right direction. I think Cloudflare's recent policy change around AI scrapers, in a similar vein, points in a similar direction. AI companies need the data and they need to...
This is a move in the right direction. I think Cloudflare's recent policy change around AI scrapers, in a similar vein, points in a similar direction. AI companies need the data and they need to pay for. How they're going to do that, given that they still aren't profitable, I don't know. But piracy is bad no matter who's doing it. If creators don't have ad revenue (in the case of websites) or sales (in the case of book authors), it makes having a healthy job market for those things virtually impossible.
Feels in line for Anthropic who at least purport to try to be a more human-friendly company. Yeah, they did the thing everybody else did, but they're also leading the charge on repercussions which sorta works.
Hopefully this leads towards actually paying for the books even if it's one license being replicated by a single company for model training, as well. I'm curious what's going to happen with the existing models.
For context, Anthropic’s valuation is $183 billion based on run-rate revenues of $5 billion.
https://www.reuters.com/business/anthropics-valuation-more-than-doubles-183-billion-after-13-billion-fundraise-2025-09-02/
I found 10 of my own works in the list. You can search here. https://www.theatlantic.com/technology/archive/2025/03/search-libgen-data-set/682094/
Enjoy the 30 grand! I don't know how much potential loss you incurred from having these works stolen but that's gotta be a pretty nice bit of news to get!
May not be that simple.
Not every book is part of the settlement just because it was in the dataset.
Nitty gritty details: https://authorsguild.org/news/what-authors-need-to-know-about-the-anthropic-settlement/
Mine were articles, not books, so I am unsure of a payout. I registered just in case, because it cannot hurt.
Can you claim that shit from Europe? I found something I wrote years ago in there!
I have no idea!
Was it published first in Europe or in the US? Are you a US citizen or no?
There's some weirdness with copyright and international treaty. I can point you towards a Bluesky thread or two, but I think if it's published in Europe first, then no, you can't. But I'm not an international copyright lawyer
Europe, published and citizen. I guess not then? If you found a good explanation online relating to that case, I’d still appreciate it, I’m curious.
EWC @ Anthropic settlement proposal for U.S. copyright claims - EWC - European Writers Council
What Authors Need to Know About the $1.5 Billion Anthropic Settlement - The Authors Guild
A number of US authors are finding out that Publishers have had a very haphazard practice about actually registering the copyrights.
If your book is registered in the US it might be good. I am trying to track down the instructions. But there's a bunch of info in the links. I believe the "first publication" part is where it gets weird for the registration.
A good person to follow on Bluesky on this is Courtney Milan and the people she talks to - she straddles authorsky and lawsky and is a lawyer who clerked for SCOTUS and has spoken on how previous settlements have violated international copyright law so we're not great at this
A thread
Update: https://apnews.com/article/anthropic-authors-book-settlement-ai-copyright-claude-b282fe615338bf1f98ad97cb82e978a1
This is a move in the right direction. I think Cloudflare's recent policy change around AI scrapers, in a similar vein, points in a similar direction. AI companies need the data and they need to pay for. How they're going to do that, given that they still aren't profitable, I don't know. But piracy is bad no matter who's doing it. If creators don't have ad revenue (in the case of websites) or sales (in the case of book authors), it makes having a healthy job market for those things virtually impossible.