It’s one thing for LLM-confabulated citations to make it into the courts or even into the academic literature. It’s something else for the primary source of information for technological civilization, Google, to rank search results highly for articles confabulated by LLMs that have undergone such cursory editing that easily-checked citations in those articles are pure confabulations.
This just happened while researching the economics of centrifugal dewatering of sargassum. Google highly ranked an article on linkedin because of one of its confabulated references:
…
The doi.org link is to an actual paper but it has nothing to do with the technology I was searching for. Moreover, going to the website for the “Journal of Environmental Management” and searching for the title turned up nothing for 2018.