Artificial intelligence developers are increasingly utilizing digitized physical books to train large language models, leading to the physical destruction of millions of printed volumes. To obtain the vast amounts of high-quality, structured text needed for AI development, specialized scanning operations purchase physical books, cut off their spines, and run the loose pages through high-speed scanners. This unbinding process permanently destroys the physical copies while converting them into digital datasets, prompting growing debate among authors and publishers regarding copyright permissions, compensation, and material waste.
- Artificial intelligence companies rely on physical books because printed literature offers dense, high-quality, and structurally consistent language data compared to general internet text.
- To digitize texts efficiently at scale, scanning facilities cut off the spines of books, allowing high-speed sheet feeders to process thousands of pages per hour.
- The destructive scanning process leaves the original physical books unusable, resulting in millions of volumes being recycled or discarded after digitization.
- Converted text is formatted into massive datasets that teach AI systems context, long-form reasoning, and specialized subject knowledge.
- The widespread digitization of copyrighted books for training AI models has sparked lawsuits and regulatory scrutiny from authors, publishers, and legal experts.
DW News is a global news TV program broadcast by German public state-owned international broadcaster Deutsche Welle (DW).
AllSides Media Bias Rating: Center
https://www.allsides.com/news-source/deutsche-welle-media-bias
Official website: https://www.dw.com
Original video here.
This summary has been generated by AI.
Legal theft under corrupt laws.
as the saying goes
evil cannot create; only corrupt and destroy
Fahrenheit 451 maxing
EU nations want to ban ALL SORTS of books and media…. While forcing you to pay public license fees for state sponsored "information"….. And then people pretend to care about some dusty hard copies.
This is YOUR OWN MEDICINE, take it!
Then they will destroy libraries. Educated people are a threat to totalitarian regimes.
This report is missing the larger point. The AI companies' entire business model is based on owning the entirety of human-produced data so they can dispense it (and its derivatives) incrementally, for a fee. They are destroying only one copy of a rare/antique book, but in the process they are claiming ownership of its contents, making all the other copies useless echoes of the original which now resides in the proprietary database of Anthropic/OpenAI. And it's shameful for the lax US laws to allow this to happen.
That's shocking…
Yet another addition to what AI done wrong.
Of course there is no evidence of unique books being destroyed … the evidence was destroyed! lol
Talk about savings felt around world as maga scam terrorism programming was dragged into street and very publicly control alti delete starting with grok tic toc x this last long.
Shame.