>LLMs are not doing searches, they are doing statistical computation of likely results. This was true of ChatGPT in 2022, but any modern platform that advertis

>LLMs are not doing searches, they are doing statistical computation of likely results. This was true of ChatGPT in 2022, but any modern platform that advertises a "deep research" feature provides its LLMs with tools to actually do a web search, pull the results it finds into context and cite them in the generated text.