The discussion revolves around the utility of Deep Research tools in conducting comparative analyses, particularly in municipal compensation studies. While the tools provide valuable new sources and streamline the collection of public data, users still find the necessity for manual verification due to inaccuracies and incomplete information. Commenters emphasize that these tools, while impressive, should not be relied upon for quick conclusions, noting the inherent need for human oversight in verifying data's accuracy. The sentiment reflects cautious optimism about future improvements in AI research tools, suggesting that true autonomy in reliable research is still a considerable way off, but improvements will come as these technologies are refined. There is a clear call for better source control to enhance reliability, with some advocating for user-driven alternatives. Overall, the sentiment conveys the need for a collaborative approach between AI tools and human researchers to ensure quality and reliability in outputs.