New framework proposes governed state for accountable clinical AI

This paper introduces a framework for accountable longitudinal clinical reasoning in AI systems, distinguishing between "generated context" and "governed state." It proposes that success in clinical AI hinges not on LLM fluency but on the governance of patient state over time. The authors outline four information requirements for accountability: an immutable evidence ledger with versioning, a distinct belief state, an observation-process model, and claim-level causal typing. They also present a six-level maturity framework, noting that current LLM-centric practices exhibit high capability but low maturity. AI

IMPACT Proposes a new conceptual framework for improving the accountability and reliability of AI systems in longitudinal clinical reasoning.

RANK_REASON The cluster contains an academic paper detailing a new framework for AI in clinical reasoning. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• Clinical artificial intelligence
• Large language models
• Victor Lorena De Farias Souza

• paper
• safety

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →