Secondhand booksellers across the UK and Ireland are reporting a surge in strange, large-scale book orders. Shop owners describe receiving requests for collections that lack any logical theme, ranging from obscure 1980s magazines to international translations of fiction. These orders are consistent in their erratic nature and high volume, leading many in the trade to suspect that artificial intelligence firms are behind the purchases.
Industry experts suggest these companies are acquiring physical books to scan their text for model training. One bookseller noted that the buyers are paying full price and show no interest in typical wholesale discounts, signaling a specific and urgent need for the raw data trapped in these printed pages. Some of the orders are traced to freight warehouses near major transport hubs, further fueling speculation about mass data acquisition.
This trend is not limited to the British Isles. Booksellers in the United States, Australia, and parts of Europe report similar experiences, creating a global phenomenon that is disrupting the traditional secondhand market. While some of these volumes are public domain, the preference for pre-2022 texts suggests a desire for data unpolluted by machine-generated content.
Reports confirm that major AI developers have spent significant capital acquiring books to slice, scan, and eventually recycle. For small shop owners, the activity is a source of both curiosity and concern as their inventory is redirected toward training the next generation of language models.