Urgent.News

One page, thousands of outlets. See who else covered it.

Editions

AI

Now Amazon is destroying rare books to train its AI.

Per a 404 Media report published this morning, Amazon, that evil empire, “is buying massive quantities of books, scanning them for AI training data, and destroying them in the process.” This news follows slow-creeping revelations that other AI companies have

Now Amazon is destroying rare books to train its AI.

Amazon is buying and scanning massive quantities of rare books for AI training data, subsequently destroying them in the process. This practice, dubbed "destructive scanning," follows similar revelations from other AI companies like Anthropic, whose "Project Panama" involved scanning all books in existence. The Washington Post broke the story, revealing that Anthropic's "Project Panama" was secretly scanning books to teach AI models to write well, as opposed to mimicking low-quality internet language.

The data is valuable because many texts are not readily available on the internet and are often free of AI-generated text, which could degrade AI models through a process called "model collapse." Amazon had previously denied practicing "destructive scanning," but new evidence suggests otherwise. Employees at an Amazon warehouse in Las Vegas, Nevada, confirm that they receive shipments of printed books, cut the bindings off, and scan the books, destroying them in the process.

Despite the environmental concerns, the practice has faced some backlash from the American public, leading to calls for accountability from big corporations.

Written by urgent.news from Literary Hub's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at lithub.com →

More in AI

Shadow Traffic Is the Only Honest Free Model Evaluation

The demo replay played in the dim glow of a conference room, and everyone agreed the free model sounded good. The same four prompts produced crisp summaries, polite error handling, and no obvious…

  • Shadow traffic approach compares free model with primary model
  • Real production traffic used to test free model's performance
  • Metrics like error rates, latency, token consumption analyzed

More from Monday 17 August →