←Back to Home

Research Frontier

Foundational
Thinking

Papers, research, and ideas from the frontier of health intelligence.

HealthMemory Papers

HealthMemory Papers•001•February 2026•8 min

Canonical Health State

A Systems Primitive

The missing primitive: a canonical, individual health state over time.

Read→
HealthMemory Papers•002•May 2026•9 min

Authority, Consent, and Individual-Anchored Truth

Consent as Canonical Assertion

Consent modeled as state, not policy. Authority becomes inspectable, revocation becomes causal, and historical truth becomes independent of present permissions.

Read→
HealthMemory Papers•003•May 2026•9 min

Time, Irreversibility, and Longitudinal Integrity

Temporal Ordering as Structural Constraint

Why append-only temporal ordering is a structural requirement, not a design preference. Longitudinal integrity is achieved through refusal to rewrite.

Read→
HealthMemory Papers•004•June 2026•6 min

Language, Interpretation, and Projection Layers

Language as Projection, Not Substrate

Canonical state is language-independent. Interpretation proposes meaning; it does not assert truth. HealthMemory does not reject language. It refuses to depend on it.

Read→
HealthMemory Papers•005•June 2026•7 min

Reasoning as a Derived Consumer

Intelligence Downstream of Truth

Reasoning is not state. It is a consumer of state. Confining reasoning to derived consumption preserves truth while enabling unlimited intelligence downstream.

Read→
HealthMemory Papers•006•June 2026•8 min

System Robustness Under Optimization and Adversarial Pressure

Minimalism as Structural Defense

Pressure is the operating condition, not the exception. Minimalism is not aesthetic. It is defensive. HealthMemory is designed to survive pressure, not eliminate it.

Read→
HealthMemory Papers•007•June 2026•7 min

Failure Without Canonical State

Observable Consequences of Structural Absence

What breaks when there is no canonical state. Truth fragments, authority migrates silently, automation amplifies distortion. Canonical state is a prerequisite for stability.

Read→
HealthMemory Papers•008•June 2026•8 min

Inference Under Mutable State

The Limits of Learning Without Stable Ground

Inference over mutable state is undefined. Canonical state does not improve inference. It makes inference possible. Systems learning without stable substrate converge, but not on reality.

Read→

Artifacts

Architecture•Coming Soon

Compounding Intelligence

How memory layers enable clinical and population-level insight without institutional extraction.

Not released
System Design•Coming Soon

The Consent Model

Designing contribution mechanisms that preserve sovereignty while enabling collective intelligence.

Not released
Research•Coming Soon

On Context Windows

Why intelligence without persistent memory is fundamentally limited.

Not released
Infrastructure•Coming Soon

The Network Primitive

Building infrastructure that treats individuals as the atomic unit of health data.

Not released
Analysis•Coming Soon

Beyond Records

Why the EHR paradigm fails the continuity requirement.

Not released

There is always more to build.