healthcare Ai - Institute for Systems Integrity

paper

Distance Distorts Signal in AI-Assisted Triage : Why Governance Fails Before the Algorithm Does

An ISI paper examining hidden governance risks in AI-assisted triage, including algorithmic drift, classification error, signal distortion, workflow workarounds, and leadership blind spots in healthcare systems.

paper

From Human-in-the-Loop to Human-with-Agency : Why AI Oversight Fails When Humans Are Present but Powerless

“Human-in-the-loop” AI often creates the illusion of oversight without real control. This ISI paper introduces the Human Agency Framework, examining why meaningful AI governance depends on preserving human judgement, authority, and intervention capability under pressure.

papers

Micromanagement as a Governance Failure Mode : Why control concentrates risk instead of reducing it

Micromanagement is not a leadership flaw. It is a governance failure. When control concentrates decision-making, organisations become fragile. This paper examines how micromanagement signals breakdowns in decision architecture, capability, and system integrity.

paper

🏛️ When AI Writes the Discharge Summary : A Governance, Duty, and Systems Integrity Challenge

Generative AI can produce discharge summaries with impressive completeness and fluency. Emerging evidence shows persistent safety risks, reframing discharge automation as a governance and systems integrity challenge rather than a productivity solution.