I would like to be able to do two things:
Use ChatGPT to do a better job of grouping similar documents, something more than just a word score.
Use ChatGPT to write a memo/report using a designated set of DT folders and using my prompting.
Both of these functions would seem to be a no-brainer as they would only require that the plugin accept a given set of DT folders. I don’t see any reason why that couldn’t be implemented straight away.
This is just @bluefrog’s personal opinion, not the one of the company. But as usual we don’t make any public announcements what’s upcoming or planned.
Dropbox have announced they are going to engage in IA and have fired some non-willing-update developers because of that.
I complained to a friend of mine, hardcore dropboxer he, saying that Dropbox is a File Management Company, and they must be centered in that and forgot in “cancamusas”, and he shut-up me extremely fast: “Rafa, think what Dropbox will discover from your stored files, relations between documents, summaries, etc.”.
He is spot-on.
Dropbox, Box, and other cloud document vendors are in a fantastic position to leverage AI technology. From my trials so far working with AI on large documents using my own scripts and API integrations, the speed of AI response is a limiting factor at present. But if Dropbox or Box were set up to summarize or do other AI functions in the background then the speed becomes much less of an issue - if I can upload it today and then in 2 days or next week when I access it the software has added summary info, then that is painless and a huge value-added feature. If this becomes a major distinguishing feature of theirs so they perfect the AI algorithms for me to make them more time-efficient, then that just makes it even more enticing and adds to the ROI.
Box in particular plans to reveal more details tomorrow at a webinar but says they will be offering enterprise-grade, HIPAA-compliant AI. If that means I can upload highly confidential medical and/or legal documents with confidence and as a plus their system will index, summarize, and/or link documents based on algorithms I assign with hyperlinks back to the raw source material - then that could be a stunning gamechanger for professionals of any type for whom document analysis is a primary part of their business.
As you say, it’s the server side processing that makes that possible. I spent the whole weekend playing with langchain, both on the OpenAI api and using local models, and it’s just not there yet for the type of use I want to put it to. It will be, but the future isn’t going to be evenly distributed. It never is. In the meantime, pressuring the devs here seems a little silly; I’m sure they are watching this space.
Also, I don’t want Dropbox or Box to put my documents in a vector db, HIPAA compliant or no.
I think much simpler things are already possible, like translation and summarization of short documents or editing suggestions on short documents.
I suspect that AI on a professional scale - like summarizing a 1,000-page document with page-specific links to the source - will always be the domain of an enterprise-grade web app but integration with DT3 will be possible.
more than possible
Not being a native speaker: what does „more than possible“ mean, exactly? „Certainly“? Or something else (but what)? And if it’s „certainly“, why is „more than possible“ preferable over it?
Not sure exactly what he meant… but specifically with regard to a question of English word choice each of these is increasingly more likely:
Possible
Probable
Certain
(For expert testimony in most courts in USA something <50% likely is possible, something >50% likely is probable, and something 100% likely is certain.)
It a way of saying “yes, and…” gesturing at something more than mere possibility. In this case, I would say that for applications within the scope a limited context window and with some degree of adult supervision the results are eerily good. For example you can feed it a 2-3 page complex form full of data laid out in table cells and just…chat with it about the contents. That’s amazing.
I just get frustrated by the extent to which reactions tend to cluster at the extremes; it’s amazing tech with complex and significant limitations, the nature and scope of which are literally changing from one day to the next.
Agreed
And even if it has limits now, it is surely worth learning because it won’t be long before limitations are overcome and it is ready for production use.
A bit off topic, but I think relevant. Have you read Jack McDevitt Academy or Alex Benedict novels? He describes conversational IA near as ChatGPT answers and builds conversations. Each spaceship, home or office has an IA that oversees repetitive tasks, and you can talk with it in the same way you do with ChatGPT plus, “please call Whatever”, “make an appointment for tomorrow” or “you have a call from Whatever. Do you want to attend?”
I think that is the future of real IA. Read one of their books, they are in series but can be read almost independently.
It is so realistic that in one of the Alex Benedict novels, burglars enter in Alex home and IA didn’t call police. Then Alex asks: “Why didn’t you call the police?”, and IA (Jakob, I think it is named) answers: “Well, they told me not to call the police because they were friends of yours” (or any similar answer, I don’t remember well, but typical to cheat an IA).
It’s just for emphasis. Implies that it’s not only possible but we have more than the minimum to do it (in the opinion of the author of course).