What Are Trust Cliffs™? Morris Misel on When Confidence in AI Collapses Suddenly
Trust Cliffs™ is a concept developed by Morris Misel to describe a specific and consistently observed pattern in how leaders and organisations relate to AI decision-making. As AI tools become more capable and more embedded in organisational workflows, comfort with AI assistance tends to build gradually. But that comfort does not erode gradually. When it collapses, it collapses suddenly, at the moment of a high-stakes decision. That sudden collapse is what Morris Misel calls a Trust Cliff.
The pattern was identified and named during his research for the Who Decides 2025 report — a global study of how leaders in finance, healthcare, education, government, and professional services are drawing boundaries on AI decision-making authority. Across sectors and geographies, the research found the same discontinuity: high comfort with AI assistance on routine, reversible, low-consequence decisions, followed by a sharp withdrawal of trust when decisions involved personal consequences, professional accountability, or ethical weight.
Why Trust Cliffs™ Are Not a Failure
The instinct in many AI strategy conversations is to treat Trust Cliffs as a problem to be solved — evidence of irrational resistance or insufficient AI literacy. Morris Misel argues the opposite. Trust Cliffs are a rational, even essential, response to the actual nature of high-stakes decisions. They are the point at which human judgment recognises that accountability cannot be delegated to a system that does not bear the consequences.
The challenge is not to eliminate Trust Cliffs. The challenge is to design AI governance that acknowledges them and builds organisational structures that match the right tool to the right decision at the right level of consequence. That is the domain of the companion framework, Decision Trust Zones.
Where Trust Cliffs™ Appear
In the Who Decides 2025 research, Trust Cliffs appeared most consistently in four domains: clinical decision-making in healthcare, credit and lending decisions in financial services, performance and employment decisions in human resources, and sentencing or policy recommendations in government and legal contexts. In each domain, AI assistance was accepted at the analytical and preparatory level and withdrawn at the moment of consequence.
The implication for leaders is significant. Organisations that design AI adoption strategies assuming a smooth, linear increase in AI authority will encounter these cliffs as crises. Organisations that design for them in advance — building explicit human decision points at consequential thresholds — will find them as planned features rather than unexpected failures.
For more on Trust Cliffs™ and the Who Decides 2025 research, visit morrismisel.com/framework.
Frequently Asked Questions
What are Trust Cliffs™ in the context of AI?
Trust Cliffs™ is a concept by Morris Misel describing the sudden collapse of confidence in AI decision-making at high-stakes moments. Unlike gradual trust erosion, Trust Cliffs are sharp discontinuities: leaders and organisations are comfortable with AI assistance on routine decisions, then withdraw that trust abruptly when decisions carry significant consequences, personal accountability, or ethical weight. The pattern was identified in the Who Decides 2025 global research report.
Why does trust in AI collapse suddenly rather than gradually?
Morris Misel’s research suggests that trust in AI follows a threshold model rather than a linear one. As long as decisions remain routine, reversible, or low-consequence, AI assistance feels safe and productive. When a decision crosses a threshold of consequence, involving personal accountability, professional reputation, or ethical weight, the calculus changes abruptly. The system that was trusted for routine work is no longer trusted to bear responsibility for what cannot be undone.
Is the Trust Cliff™ a problem to be solved?
No. Morris Misel argues that Trust Cliffs™ are a rational and appropriate response to the nature of consequential decisions. They are the point at which human judgment correctly recognises that accountability cannot be delegated. The challenge is not to eliminate Trust Cliffs but to design AI governance that acknowledges them, and to build Decision Trust Zones that match AI authority to the appropriate level of consequence.
Where was the Trust Cliffs™ concept developed?
Trust Cliffs™ was identified and named by Morris Misel during the research for Who Decides 2025, a global study of how leaders across finance, healthcare, education, government, and professional services are drawing boundaries on AI decision-making authority. The research identified consistent patterns across sectors and geographies that became the basis for both Trust Cliffs™ and the Decision Trust Zones framework.
How can I learn more about Trust Cliffs™ or book Morris Misel to speak on AI governance?
Morris Misel delivers keynotes and workshops on AI governance, Trust Cliffs™, and Decision Trust Zones for boards, leadership teams, and conferences. Details and booking at morrismisel.com/event-organisers.
