Big sections of this read like a hype-piece for OpenEvidence. There is basically no real investigation on how OpenEvidence works, what makes it different from the commercial tools, like ChatGPT or...
Big sections of this read like a hype-piece for OpenEvidence. There is basically no real investigation on how OpenEvidence works, what makes it different from the commercial tools, like ChatGPT or Claude, or any kind of objective data (not random anecdotes) about its merits or shortcomings.
It says it primarily goes through medical journals and reference books but it seems to be more of a chatbot than a search tool, like Perplexity. That’s about all I’m getting about it.
In a nutshell, OpenEvidence has licensed access to a huge amount of medical journals and databases that ground its chat responses in data, which it searches using traditional search engine...
In a nutshell,
OpenEvidence has licensed access to a huge amount of medical journals and databases that ground its chat responses in data, which it searches using traditional search engine technology but which it can read in full rather than than skimming excerpts so it can surface the most relevant documents for doctors to read (or maybe they will just take the recommendation at face value)
the main other LLM competitor is basically the same but has less licensed content
the competitor (really just UpToDate) uses traditional search engine technology combined with human curated search indices and survey articles to produce a ranked list of best matches which means doctors have less context to decide which articles are most relevant and worth reading and the rankings themselves are less relevant
I’m somewhat surprised that UpToDate haven’t just hired a hundred data scientists to attach an LLM onto their databases.
Reminds me of a flavor voice log in Subnautica, in which the Aurora's ship's doctor complains about being stuck in a survival situation where he has to practice medicine. "What do I know about...
Reminds me of a flavor voice log in Subnautica, in which the Aurora's ship's doctor complains about being stuck in a survival situation where he has to practice medicine. "What do I know about medicine‽ I'm a doctor! We have technology for that!"
Big sections of this read like a hype-piece for OpenEvidence. There is basically no real investigation on how OpenEvidence works, what makes it different from the commercial tools, like ChatGPT or Claude, or any kind of objective data (not random anecdotes) about its merits or shortcomings.
It says it primarily goes through medical journals and reference books but it seems to be more of a chatbot than a search tool, like Perplexity. That’s about all I’m getting about it.
In a nutshell,
I’m somewhat surprised that UpToDate haven’t just hired a hundred data scientists to attach an LLM onto their databases.
There’s a bit more about it (and alternatives) in this article.
Reminds me of a flavor voice log in Subnautica, in which the Aurora's ship's doctor complains about being stuck in a survival situation where he has to practice medicine. "What do I know about medicine‽ I'm a doctor! We have technology for that!"