A recent study examined the capability of large language models (LLMs) to classify political content from URLs across multiple countries, specifically focusing on news articles from France, Germany, Spain, the UK, and the US. While LLMs have demonstrated promise in labeling tasks, their effectiveness in accurately distinguishing political from non-political content remains largely untested. The research highlights the challenges these models face when applied to diverse media landscapes and political contexts. By analyzing content from various global sources, the study aims to assess whether LLMs can reliably identify political material beyond simple keyword detection or superficial cues. This investigation contributes to ongoing efforts to understand the limitations and potential of AI-driven content classification in the realm of political news. The findings underscore the need for further evaluation and refinement of LLMs to improve their accuracy and applicability in global media analysis.