> can take hundreds of pages of text and distill it down to an executive summary of any length
This is the kind of breathless claim that no doubt fuels the skeptics.
None of the context windows are large enough for "hundreds of pages" nor "executive summaries of any length".
I did believe that it's possible to make LLMs do that kind of task with significant engineering effort to make it do summaries iteratively somehow, essentially "compressing" parts of the document recursively. But it's not something that you can just give to ChatGPT and have it work.
So yes, the hype is real, in both ways: there's lots of potential to explore over the next years, but also a lot of the claims you read today are not sufficiently hedged, which makes them look outlandish.
This is the kind of breathless claim that no doubt fuels the skeptics.
None of the context windows are large enough for "hundreds of pages" nor "executive summaries of any length".
I did believe that it's possible to make LLMs do that kind of task with significant engineering effort to make it do summaries iteratively somehow, essentially "compressing" parts of the document recursively. But it's not something that you can just give to ChatGPT and have it work.
So yes, the hype is real, in both ways: there's lots of potential to explore over the next years, but also a lot of the claims you read today are not sufficiently hedged, which makes them look outlandish.