C. C. Terence Tan, Dinesh Visva Gunasekeran, Cheng Ooi Low, Gabrielle Sze Yee Sim, Danella Yaoxin Foo, Robert J. T. Morris, Tien Yin Wong
Clinical vibe coding needs risk-stratified governance now, before largely invisible adoption outpaces the field's capacity to assess it. VIBE-HI offers an architecture institutions can apply immediately and provides a clear pathway for empirical validation, beginning with a modified-Delphi consensus study and stakeholder review.
In a recent article, Weissman et al. 1 examined the extent to which artificial intelligence (AI)-based large language models (LLMs) generate clinical decision support (CDS) outputs that meet the criteria of regulated medical devices 1 and called for new regulations for LLM-based CDS systems. In this manuscript, we respond to the proposed considerations highlighting those that have been addressed by existing guidelines and would not need new frameworks, as well as the need for new regulations of “generalized” CDSS that are not anchored to specific clinical indications. We contextualise this regulatory gap with an overview of the literature distinguishing between confined and unconfined AI systems. We also outline specific areas in which new regulations may be required, along with risk mitigation strategies that could be incorporated in new guidelines.