Microsoft Executive Warns of Low-Quality AI Content Polluting Workplace Communications
Microsoft Executive Vice President Ryan Roslansky warned that low-quality AI-generated content ("AI slop") has invaded daily office communications, creating a vicious cycle where AI generates, reads, and summarizes text. To combat this issue on its platform, LinkedIn introduced a "Seems like AI slop" reporting option and deployed automated classifiers to filter out low-quality AI spam. As enterprises rapidly adopt generative AI tools, over-reliance on unverified text threatens to dilute real insight and compromise critical decision-making across workplace workflows. This highlights the urgent need for corporate governance guidelines and human review mechanisms to preserve the value of enterprise communications. Roslansky emphasized that while generative AI accelerates drafting and data organization, it cannot guarantee factual accuracy, logical completeness, or original perspective. He advocated for mandatory human review, data source verification, and explicit ownership rules before sharing AI-assisted documents downstream in business operations.
## BACKGROUND
"AI slop" refers to low-effort, low-quality digital content generated automatically by artificial intelligence tools without adequate human oversight. As generative AI tools became widely accessible, users increasingly relied on them to quickly draft emails, reports, and social media posts, leading to an influx of generic text online and in corporate environments. Consequently, platforms and enterprise systems are deploying automated classifiers and user feedback tools to detect AI-generated spam and safeguard content quality.