Integration of Evidence and Ask Heidi is NOT GOOD - please go back to separate!
under review
A
Abbie Mullins
PLEASE go back to keeping evidence and Ask Heidi separate. This new integration of "chat" feature with both evidence and asking Heidi is slow, doesn't give me the output needed, cannot keep up if I am asking Heidi to add something to one note output and then a question about something else in this new "thread" style. It is much less helpful, productive, and doesn't give me what I need in a timely manner. Keep evidence in one area, and this ask Heidi feature for generating new notes/documents or working within the notes from our current visit. Your upgrades are always great and this one truly misses the mark and has made my work more frustrating and longer. Please go back!
C
Charlotte Ballisager
Yes, this is horrible. When I ask Heidi a specific question or to list certain tasks, it initially creates the answer, but then it disappears behind a long scroll of information I do not need, such as evidence, sources, and incorrect assumptions of treatments and goals.e What I wanted to add to my note is lost behind the scroll. This is particularly problematic when editing in my note, and I lose the edit. Personalization does not seem to work, and using the AI does not save changes even though I follow the instructions. Please offer to turn off the new AI as it was working well before for my purpose. Essentiall the problem is that the AI gives information that is not requested and tends to get into diagnostics.That is a huge problem.
Anya Sharma
marked this post as
under review
Tom
Hey Abbie Mullins — thanks for spelling this out, and sorry this change has made things slower and more frustrating. Super helpful feedback. Just to make sure we’re looking at the right thing: is the main issue the new combined “thread” flow (hard to switch between editing a specific note output vs asking a separate question), the speed/latency, or that the responses themselves are worse? Any concrete example of what you were trying to do (e.g. “add X to this note output, then ask Y about something else”) would help us reproduce it. We’ll dig into it.