🚩 Report: Illegal or restricted content
this dataset was scraped from Cara, a website that explicitly disallows AI data scraping in it's TOS. This infringes the TOS entirely and is illegal as such.
This dataset also infringes directly on the protected copyrights of thousands of artists, and very few of those artists have given explicit permission to use their images in AI training.
This dataset also contains my own images that I do not grant permission to reproduce or use in any way, as all legal rights to use, reproduce, or exhibit those images belong solely to me and cara.app. that also includes AI use for which you have not been given permission.
It's not obvious in the web UI but it looks to me like this does not, in fact, contain any artwork. I believe someone is trolling you guys.
$ curl -L https://huggingface.co/datasets/CaptiveDreamer/CaraArchive/resolve/main/master_fullres_8.5m_links.txt | head
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1119 100 1119 0 0 11251 0 0
0 0 0 0 0 0 0 0 0https://cdn.cara.app/production/posts/000005f2-65cc-4702-932c-3bd1b2a7af1f/1718943637-fnZNsWgz72P51nnpCLCB9-post-cover-image.jpeg
https://cdn.cara.app/production/posts/000005f2-65cc-4702-932c-3bd1b2a7af1f/realitybean-nZPIvKUV26ShO2FMjY5Mo-1718943455166.png
https://cdn.cara.app/production/posts/000005f2-65cc-4702-932c-3bd1b2a7af1f/realitybean-i_H6DSYCl9UfAilOV-V9d-1718943455330.jpg
https://cdn.cara.app/production/posts/000006f5-a7bc-4052-af06-72bea2429da1/ivinetop-Sds1MvDbZuEGhxtv3OIz3-post-cover-image.jpeg
https://cdn.cara.app/production/posts/000006f5-a7bc-4052-af06-72bea2429da1/ivinetop-yHBMiwkpcHlSdPi0bYY4G-bule.jpg
https://cdn.cara.app/production/posts/00000798-98b7-456d-b74d-e990d4353972/taintedowl-SnuQWGN_VwdgeCJdPdLtp-post-cover-image.jpeg
https://cdn.cara.app/production/posts/00000798-98b7-456d-b74d-e990d4353972/taintedowl-AcToQ4TGTqKPUJRyIj5pU-Sunflower-man.jpg
https://cdn.cara.app/production/posts/000008c0-29be-491d-ad22-663ba786c1c0/madpxl-mXMFIcskkmqaVvJyEz1LW-ABF1B759-8472-469F-80F2-5F0939B698E0.jpg
https://cdn.cara.app/production/posts/000010c8-0434-4977-8771-1a948d3b211c/1726462421-Vdi1t109UDrGkJZmd3pAM-post-cover-image.jpeg
https://cdn.cara.app/production/posts/000010c8-0434-4977-8771-1a948d3b211c/artcostiander-DDI4I9JoKIpzZ2KsxPXuk-ECD11CEA-A002-404F-A17E-2759B1692FA7.jpg
curl: (23) Failure writing output to destination, passed 14639 returned 659
[swiley@gdp]:~ 18:05:40
$
I haven't checked the rar file but it would be very impressive if the images were somehow shorter than their links.
even if its just links, they are still links posted with specific intent to have AI trained on images for which no permission has been granted to train on. still very much illegal on account of promoting illegal activities and intent to commit illegal activities.
Scraping and posting links is not illegal in the US (AFAIK IANAL.) There's a lot of precedent for this.
still amoral. i dont give a crap if its legal, using people's art in a way they dont consent to is a bad thing to do, and arguably still infringes copyright laws that ARE well established because it is a form of reproduction, which copyright laws explicitly disallow.
i dont give a crap if its legal
this is why the world is becoming enshittified by censorship, because of people like you.
still amoral. i dont give a crap if its legal, using people's art in a way they dont consent to is a bad thing to do, and arguably still infringes copyright laws that ARE well established because it is a form of reproduction, which copyright laws explicitly disallow.
It has not been even remotely established if models are derivative works of their training data.