Researchers warn that the widespread use of AI systems could unify human expression and homogenize culture and cognition. ^[1] The paper proposes a novel technique for accurately and automatically inferring LLM-related scrapers by hosting dynamic websites that serve unique canary tokens to each visiting scraper. ^[2] Via experiments across 22 production LLM systems, the authors demonstrated that their approach can reliably identify which scrapers feed which LLM. ^[3] A study at the University of China found that the reflexive use of AI can affect students' learning capabilities, though it does not prove causality. ^[4] A study published in March compared 22 AI models and over 100 humans on creative challenges, finding that AI responses were slightly more original but more similar to each other than human responses. ^[5] Large-scale web scraping to feed LLMs can affect site stability and raise legal, privacy, or ethics concerns. ^[6] The approach can identify scrapers that are not publicly known or disclosed by the companies. ^[7]