AI companies copying internet to train models biggest theft of labour in history

19-Sep-2026

.

AI models like ChatGPT and Claude were trained on vast online data, including copyrighted content, sparking lawsuits from news organisations like The New York Times. 

Internal Microsoft documents revealed executives criticizing AI's use of such content as large-scale theft of creators' labour without consent or compensation. 

Microsoft and OpenAI defend this under fair use, claiming transformation rather than replication. 

Plaintiffs argue AI substitutes original news, reducing site visits and revenue. The case highlights tensions between AI innovation and intellectual property rights in media.