DevonThink summarization AI feature request

Reading about Elicit: Joining Ought (via a blog) a few minutes ago, I was struck by how this aspect of the “AI”:

Their main product at the moment is Elicit - an AI assistant for academic and professional researchers. It uses NLP to find research papers, synthesise them, and extract research questions, evidence, and arguments from them. They’re currently focused on helping people do literature reviews. But the plan is to expand Elicit’s capacities to help with the whole research process.

I found the results were on par with what Google Scholar or Semantic Scholar would turn up. But the Elicit results show why it returns certain papers – each paper has a GPT-3 generated summary that tries to answer the original question. It’s a small difference, but a huge help when you’re drowning in PDFs and trying to quickly find the right ones to read.

Wow. Just wow. My DT database is a morass of research papers. This level of summarization and justification, would take DT up a couple of levels for me. I will never find the time to read all of the research in my field (building effective teams). This would make it easier to find needles in the haystack.

Please consider as a feature request.


…but could also keep you trapped in a filter bubble. Anyway, thanks for the request!

1 Like

That was my first thought when I read up on Elicit.

@cgrunenberg and @Blanc a filter bubble is a risk, but it is already a risk using DT in the first place. Search will only ever find what I put into DT.

In my case I’m an omnivore - I read very widely for my work. My problem is that I sometimes don’t realize I’m missing great material that is buried in hard to read academic papers. Anything that highlights an item and it makes it clear its worth reading makes my work more effective.