Unverified50% confidenceFactExact time
Domain shift is the core mechanism behind cross-institutional generalization failures in medical AI, arising from feature distribution differences caused by different hospitals using different laboratory instruments, clinical pathways, and coding practices.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe dominant failure type in AI medical records is omission, not transcription errors.71% similarUnverifiedDeep learning AI systems experience a sharp drop in performance in scenarios outside their training data distribution (the distribution shift problem), and are less capable than experienced specialists at recognizing rare diseases.70% similarUnverified'Misalignment' refers to a divergence between the actual behavior of an AI system and the intentions or values of its human designers.70% similarUnverifiedSycophancy is a classic "outer alignment" failure case in AI alignment, where a discrepancy exists between the training objective (making users satisfied) and the true objective (being genuinely helpful to users).69% similarUnverifiedThere is a vast gap between those who use AI well in frontend design and those who do not, making a model's baseline capability more decisive for ordinary users.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1001106API
curl https://kongchang.com/api/v1/knowledge/claims/1001106MCP
get_claim(id=1001106)