Internal emails from Microsoft and OpenAI have revealed deep concern inside both companies about the impact of AI scraping on the news industry. According to the messages, a Microsoft executive described AI scraping as the largest theft of labor in human history. The emails also show fears that a doom loop could destroy news organizations, as AI systems harvest content without payment while replacing the need for readers to visit original publishers.
The disclosure is striking because Microsoft and OpenAI are currently facing lawsuits from publishers accusing them of using copyrighted articles to train AI models without permission or compensation. The emails suggest that key figures inside the companies privately understood the severity of the issue long before courts began to weigh in. Calling the practice theft signals that the internal debate was not just about legal technicalities, but about the fundamental fairness of using human writing to build automated systems.
The notion of a doom loop adds another layer of worry. If AI models continue to scrape news content and then present summaries directly to users, publishers lose traffic, ad revenue, and the ability to fund more reporting. As newsrooms shrink, the quality and quantity of original information drops, which in turn degrades the very data the AI models depend on. The loop feeds itself and could leave both journalism and AI worse off.
These revelations come as courts and lawmakers around the world are still deciding how copyright law applies to AI training data. The emails could strengthen the hand of publishers seeking payment or licensing deals. They also increase public pressure on AI labs to find ways to compensate content creators. What happens next will likely shape whether the internet remains a place where original reporting is economically viable, or whether human-written news becomes little more than free fuel for machines.