Dear Anthropic Please Stop Flagging My Questions Related to Medical Education/Board Review
Claude model activity is surfacing — worth tracking for capability changes, ecosystem impact, and availability.
A user preparing for medical board exams is frustrated with Claude's inconsistent flagging and rerouting of their medical education questions.…
Finished residency, got state licensed, now prepping for boards. Using Claude for complex med ed for some time. Has helped with through residnecy, attending onboarding, random life stuff. It knows me well.
Im studying/doing board review with QBanks and when I ask for explanations it flags me and reroutes me with no rhythm or reason. Routed randomly between sonnet and opus sometimes even haiku. The difference in output quality when it reroutes is noticeable. Weirdly, every now and then it keeps my response on its latest model of my choosing… without rerouting…randomly. The routing is not as intelligent as I thought Anthropic would make it.
I’ve sent countless feedback requests. I get that these are intentionally high tier safeguards, but come on man. I’d furnish whatever documentation they need to authenticate my credentials. I recognize I should be greateful to have what they offer, but it’s a bit unfair.. to me… other healthcare providers.. even the patients we hope to treat.
As someone with ADHD, Claude catching my thought process headed down a rabbit hole, away from the medical concept in the question, did more for me than any text explanation ever could. All of this happened way less before the latest model was taken down. I know I’m not the only one, and other domains of life/work get flagged the same way. I’m beyond frustrated.
Dear Anthropic please stop. There is nothing malicious about me using your tool to be a better. You’re going to give me a f*ckn stroke, so I must “push back”