28 stories tagged ai-training-data, newest first.
429 points by 0in | 155 comments
Nieman Lab article reporting that Meta's new commercial terms, effective October 30, let Meta scrape websites of business accounts using Facebook, Instagram, or WhatsApp.
37 points by cdrnsf | 13 comments
A personal blog opinion post on coryd.dev arguing AI companies are parasites that scrape content without compensation, train models on it, and sell metered access.
576 points by papergirl | 541 comments
Authors Guild press release on unsealed briefs in its lawsuit alleging Microsoft and OpenAI executives knew using pirated books for AI training was illegal.
788 points by pluc | 681 comments
TechCrunch article reporting that unredacted court filings show a Microsoft executive called AI scraping 'the largest theft of labor in human history.'
360 points by speckx | 212 comments
TorrentFreak article on Strike 3 Holdings seeking to tie a John Doe torrent lawsuit against a Meta Reality Labs executive to its $446M AI training case against Meta.
669 points by darccio | 394 comments
Guest post on Anna's Blog claiming AI companies buy and destroy physical books for training data, urging volunteers to scan and upload rare books.
527 points by Cider9986 | 834 comments
Anna's Archive blog guest post arguing AI companies are buying and destroying physical books for training, and urging volunteers to scan and upload rare books.
1360 points by speckx | 295 comments
Personal blog post contrasting the prosecution of RSS co-creator Aaron Swartz for scraping with Meta's large-scale scraping facing little consequence.
544 points by pseudolus | 371 comments
The Register article reporting that Google paid $10 million at auction for defunct Spirit Airlines' data, including emails and recorded calls, for AI use.
620 points by anon373839 | 372 comments
A tweet by @HedgieMarkets saying AI companies bulk-buy rare books via ISBNdb, cut off their spines to scan them, and shred the originals to obtain training data.
517 points by BeetleB | 490 comments
AP News article reporting a judge's approval of a $1.5 billion settlement by Anthropic over pirated books used to train its Claude chatbot.
373 points by spankibalt | 337 comments
Variety news article on a publishers' lawsuit, including author Scott Turow, alleging Mark Zuckerberg personally authorized Meta's copyright infringement in training AI models.
737 points by dlx | 491 comments
Reuters report, shown via Economic Times, that Meta is installing software to capture US employees' mouse movements, clicks and keystrokes to train AI agents.
554 points by kevcampb | 124 comments
News article reporting that Atlassian has enabled default collection of customer data to train its AI models; the capture shows only a Vercel browser verification page.
552 points by vmg12 | 248 comments
Ask HN-style post warning that GitHub will use private repos for Copilot AI training unless users opt out by April 24, with a link to the settings page.
393 points by askl | 220 comments
TorrentFreak article reporting Meta's argument in a lawsuit that uploading pirated books via BitTorrent while training AI models qualifies as fair use.
904 points by acomjean | 686 comments
New York Times article on Anthropic agreeing to pay $1.5 billion to settle a copyright class action brought by book authors; the capture shows a captcha page.
383 points by pyman | 517 comments
Business Insider article on a judge's ruling that Anthropic scanned millions of cut-up used books and downloaded over 7 million pirated ones to train Claude.
369 points by stephendause | 257 comments
New York Times article on Cloudflare blocking AI data scrapers by default; the captured page shows only a CAPTCHA verification challenge.
438 points by pier25 | 254 comments
The Verge article reporting that Facebook is asking users to let Meta AI process private, unpublished photos from their camera rolls.
336 points by croes | 201 comments
The Register reports the head of the US Copyright Office was fired a day after the office's draft report found AI training on copyrighted works can exceed fair use.
1118 points by gameshot911 | 804 comments
Ars Technica article on unsealed Meta emails in an authors' copyright case alleging Meta torrented and seeded 81.7 TB of pirated books to train AI.
1184 points by latexr | 522 comments
404 Media article on OpenAI accusing DeepSeek of using its model outputs, noting OpenAI itself trained on others' data without permission.
382 points by lopkeny12ko | 209 comments
A post on X by Rachel Tobac saying LinkedIn is using users' content to train its AI tools; the post content did not load.
650 points by earthboundkid | 775 comments
A Cat and Girl webcomic in which the cartoonist reflects on being named in a court document listing artists whose work was used to train Midjourney.
407 points by alabhyajindal | 128 comments
Mozilla Foundation campaign page asking Microsoft to clarify whether its new Service Agreement allows using personal data from its products to train AI.
412 points by version_five | 329 comments
Blog post reporting that Japan's government will not enforce copyright on data used for AI training.
857 points by _delirium | 197 comments
Internet Archive blog post by Brewster Kahle describing how heavy automated scraping traffic brought archive.org down on May 29, 2023.