🚩 Report: Illegal or restricted content
The dataset is a list of links for the art website Cara. Cara's TOS states the following.
“Any attempt by you to disrupt, or encourage or promote the disruption of the Cara Site, any Service, any Art and other User Generated Content, and/or any related service may be a violation of criminal and civil laws. You agree that you will not violate any applicable law, rule or regulation in connection with your use of the Cara Site, any Service, any Art and other User Generated Content, and/or any related service nor will you interrupt or attempt to interrupt the operation of the Cara Site, any Art and other User Generated Content, and/or any related service in any way, including without limitation, by doing any of the following:
…
iii) facilitating, creating, or maintaining any unauthorized connection to the Site, any Art And Other User Generated Content, and/or any Service, and/or the unauthorized scraping of any images and/or any Art from the Cara Site without the express written permission of the Artist responsible therefore;”
While this dataset does not directly scrape the image files from Cara, the compilation of links to each piece of art into a dataset for AI training usage can be considered facilitation of scraping because the links make it significantly easier to grab the data as compared to having to run your own web scraper. As stated above, this is still against Cara's TOS.
While this dataset does not directly scrape the image files from Cara
Then it is not Illegal or restricted content.
the compilation of links to each piece of art into a dataset for AI training usage can be considered facilitation of scraping because the links make it significantly easier to grab the data as compared to having to run your own web scraper.
The U.S. Ninth Circuit Court of Appeals ruled in the hiQ Labs v. LinkedIn case that scraping publicly accessible websites does not violate the Computer Fraud and Abuse Act (CFAA), even if it violates the website’s Terms of Service. Scraping is not illegal. "Facilitation of scraping" is not illegal.
The scraping that is being facilitated is that of copyrighted images, which the ruling doesn't apply to. If scraping the images is a violation of copyright then it would be facilitation of copyright infringement which is illegal.
The scraping that is being facilitated is that of copyrighted images, which the ruling doesn't apply to. If scraping the images is a violation of copyright then it would be facilitation of copyright infringement which is illegal.
The ruling states that there is to be no differentiation from a bot and a browser. Just take the L