Thank you @Anwar Butt for your reply. I did ask Claude how to recover the chat and/or the project but it references itself and can't troubleshoot itself, which is a bit of a pain. HOWEVER, I eventually found a workaround! From the chat list (which still showed all my chats, including archived ones that I could not retrieve), I clicked the three-dot menu (⋯) next to the chat I wanted to recover and selected "Move to another project." Simply doing that was enough to make the chat accessible again. No data was lost.🥳

@Dianne Martin stopping here only to say that your job sounds super cool 😎

@Pam-Larry Ratz for handling multiple research papers I found NotebookLM phenomenal. You can upload up to 50 sources in different formats - pdfs and websites for example - and tell the LM to summarise, create a report, or a presentation or to test you on the content and much more. For each output you can give additional instructions to tailor further. I did some testing with research I am super familiar with before using it for new stuff and the output was mind blowing. I have been able to do days of work in a couple of hours. Have a go, you won't look back!

I realise that i am in the minority here but I do not use my Ps and Qs with AI for a couple of reasons. First AI is an instrument, a tool, not a human. I don't say thank you to Google Scholar for finding research articles and I don't say thank you to my electric toothbrush for my shiny white teeth. Likewise, I don't say thank you to Chat GPT for the many thinks it helps me achieve. In fact, I think there is an intrinsic risk in talking to a machine as I would to a human. Second, there are lots of articles on the Internet about the environmental cost of saying please and thank you to Chat GPT, which is another reason for me not to. I How saying 'please' and 'thank you' to ChatGPT costs OpenAI millions | The Standard https://share.google/B5TycwvTEDKEZ5Jyd