With the release of ChatGPT API, plenty of work can be assisted with AI, including file management in DEVONthink, of course.
I have come up with a GPT-based workflow for academic research. Currently, ChatGPT has the excellent language capabilities, including natural language understanding and writing. However, what it lacks is the ability to observe and infer. So it would be great if I, as a researcher, can feed ChatGPT with a lot of specific materials which are saved in DEVONthink database, and then ChatGPT can learn them and return summaries and retrieve specific information I asked, of course, with corresponding sources for citation.
A month ago, I thought it was impossible in the current period, but now I think this workflow could come true. I really hope that ChatGPT API can be integrated into DEVONthink
There is a similar online websites that can do this kind of work, but I hope it can be achieved locally and across multiple PDFs; i.e. my DEVONthink database. Do you guys have any ideas?
You don’t need to wait…. Use applescript to feed the data to chatGPT already. Devonthink is perhaps one of the most ideal macOS apps out there in how well they let you get your data and use it - including making your own integration like this at any time you want it.
I just want to chime in here after trying out the website you linked @TomBen. After uploading a couple PDFs, I find the functionality of asking questions and getting answers based on the PDF’s contents super useful.
This kind of functionality in DevonThink would add a lot of value in my view. Especially if, as I believe the OP suggests, the questions could refer not just to one PDF but to a whole collection.
