Risk boundaries: where to draw the line on longevity interventions

Risk Boundaries

This file defines the risk boundaries that protocol design cannot cross in this repository.

Its purpose is not to eliminate risk entirely. Its purpose is to make clear where protocol logic must stop, slow down, or refuse escalation.

A protocol becomes dangerous when its risk boundaries are vague.

This file exists so they are not vague.

Core Position

Risk is not an afterthought in protocol design.

Risk is part of protocol quality.

An intervention is weaker, not stronger, when:

biological risk is unclear
interpretation risk is ignored
translation burden is hidden
implementation burden exceeds likely benefit
functional reality is sacrificed to molecular ambition

That means a protocol is not judged only by what it might do if everything works.

It is also judged by what it risks doing when things do not.

Why This File Matters

The repository has already established that:

the biomarker problem is structural
function takes precedence when biomarkers conflict with it
the strongest current interventions are foundation-level, not frontier-level
several of the most exciting interventions remain exploratory because their risk boundaries are still unresolved

That means protocol design cannot be built only from upside, or even from upside plus evidence.

It also has to be built from explicit limits.

Without risk boundaries, escalation becomes arbitrary.

What a Risk Boundary Is

A risk boundary is the point at which protocol logic should stop treating an intervention as acceptable for structured use.

That point may be reached because of:

biological danger
poor interpretability
function worsening
burden accumulation
recovery failure
population misfit
implementation breakdown
unresolved uncertainty that is too large for the intervention’s role

A risk boundary is not the same thing as total prohibition.

It is the line where protocol confidence should no longer continue as if the intervention still fits.

Main Risk Domains in Protocol Design

1. Biological Risk Boundary

This boundary is crossed when the intervention creates too much plausible biological harm relative to its benefit.

Examples include:

oncogenic risk
dedifferentiation risk
tissue toxicity
immune suppression
healing impairment
excessive metabolic strain
overtraining or recovery collapse
destabilization of organismal regulation

If the intervention’s biological downside is too poorly bounded, it should not move upward in protocol importance.

2. Functional Risk Boundary

This boundary is crossed when function worsens, resilience narrows, or recovery fails.

Examples include:

reduced capacity
rising fatigue without adaptation
reduced recovery quality
increased frailty burden
worse physical performance
worsening day-to-day functioning
a protocol that looks cleaner molecularly while the organism feels and performs worse

This is one of the strongest boundaries in the repository.

If function worsens meaningfully, the protocol should not be defended by molecular optimism.

3. Interpretation Risk Boundary

This boundary is crossed when the intervention can no longer be interpreted honestly enough to guide structured use.

Examples include:

biomarkers moving in contradictory ways
function and biomarkers diverging without a credible explanation
too much signal ambiguity
too much dependence on speculative meaning
more mechanistic language than real clarity

An intervention that cannot be read clearly enough should not be escalated simply because it remains interesting.

4. Translation Risk Boundary

This boundary is crossed when the intervention may be biologically serious, but too fragile, too narrow, too burdensome, or too infrastructure-dependent to justify protocol use.

Examples include:

complex delivery requirements
repeated procedural burden
dependence on rare infrastructure
weak fit outside ideal trial conditions
poor feasibility for longitudinal use
narrow disease-context dependence disguised as general-use logic

This boundary matters because some interventions fail not biologically, but operationally.

5. Implementation Risk Boundary

This boundary is crossed when the protocol becomes too hard to hold well.

Examples include:

adherence collapse
stacking too many interventions
recovery burden becoming too high
dose or schedule complexity overwhelming the user
foundation quality deteriorating because too much is being added
support interventions becoming burdensome enough to stop being supportive

A protocol that cannot be lived is weaker, even if it is elegant on paper.

6. Population-Fit Risk Boundary

This boundary is crossed when the intervention is being used outside the population where its logic remains credible.

Examples include:

treating a high-burden intervention as if it fits healthy adults equally well
importing disease-context evidence into general-aging protocol logic without enough support
ignoring frailty, age, baseline burden, recovery capacity, or metabolic fit
assuming one dosing logic works across all bodies

A protocol that ignores fit is crossing a risk boundary even before obvious harm appears.

Absolute and Relative Boundaries

Some boundaries are closer to absolute. Some are relative.

Absolute-style boundaries

These are boundaries where crossing them should almost always stop escalation.

Examples include:

clear biological instability
clear functional decline
strong evidence of unacceptable harm
direct contradiction of the foundation layer
protocol structures that require pretending function does not matter

Relative boundaries

These are boundaries where context, population, burden, or intervention role matters.

Examples include:

moderate side-effect burden
implementation difficulty
incomplete biomarker alignment
narrower translational fit
support-layer interventions that help some people more than others

The repository should distinguish between these, rather than flattening every risk into one category.

Risk Boundaries by Protocol Layer

1. Foundation Layer

The foundation layer has the widest tolerance for imperfect mechanistic certainty and the narrowest tolerance for functional harm.

That means:

function worsening is a major stop condition
adherence collapse is a real risk boundary
recovery failure matters
overcomplication that weakens the base crosses the line

The foundation layer should remain the most stable and least fragile part of the protocol.

2. Support Layer

The support layer has a narrower tolerance for burden and ambiguity.

A support intervention crosses its risk boundary when:

it stops supporting the foundation
it adds more burden than relief
it becomes biomarker decoration
it complicates adherence or recovery
its benefits become too narrow or too speculative relative to added load

Support must remain support.

3. Exploratory Layer

The exploratory layer has the narrowest tolerance for promotion and the widest tolerance for scientific uncertainty, but only as long as that uncertainty is explicit.

An exploratory intervention crosses its boundary when:

it is talked about as if it were protocol-ready
biomarker movement is used to bypass weak function
its risks are softened or hidden
its scientific importance is used as a substitute for readiness

Exploratory does not mean safe to drift upward by tone alone.

Boundary Conditions for Escalation

An intervention should not escalate in protocol weight if any of the following are true:

function worsens
recovery worsens
burden becomes unsustainable
risk becomes clearer and more serious
biomarkers and function diverge without a credible reason
the intervention begins displacing the foundation
the protocol becomes harder to interpret than to justify

This repository should prefer stalled escalation over dishonest escalation.

Boundary Conditions for De-escalation

An intervention should move downward, pause, or be removed when:

its function case weakens
its burden rises
its fit becomes worse for the intended population
the implementation cost becomes too high
its support role becomes decorative rather than useful
it begins requiring too much explanation to defend continued use

De-escalation is not failure. It is protocol honesty.

How Biomarkers Interact With Risk Boundaries

Biomarkers can help detect risk boundaries, but they do not erase them.

Biomarkers are useful when they show:

rising burden
biological mismatch
stress accumulation
poor adaptation
disagreement between theory and organism

But biomarkers cannot rescue an intervention that has already crossed a functional boundary.

If function declines, a favorable biomarker does not cancel that fact.

This rule is absolute in this repository.

How Combinations Interact With Risk Boundaries

Combinations deserve extra caution because they can cross boundaries more quietly than single interventions.

A combination may cross a risk boundary when:

total burden becomes too high
attribution becomes impossible
recovery cost rises
one intervention undermines another
the stack becomes harder to live than to explain
molecular gains disguise organismal strain

Combination sophistication does not excuse boundary crossing.

Practical Risk-Boundary Questions

Before any intervention is allowed to remain or rise in protocol structure, the repository should ask:

What is the main biological risk here?
What is the main functional risk here?
What is the main interpretation risk here?
What is the main implementation risk here?
What population-fit limit matters most here?
What would count as a clear stop condition?
What would count as too much burden for too little gain?

If those questions cannot be answered clearly, the protocol is not bounded well enough.

Relationship to the Rest of the Repository

This file is directly constrained by:

03_INTERVENTIONS/12_risk_hierarchy_and_translation_limits
because that file established the repository-wide risk logic

04_decision_rules
because decision rules require explicit stop and slowdown conditions

05_biomarker_use_in_protocols
because biomarkers may detect risk, but cannot overrule function

06_function_first_logic
because function remains the strongest boundary signal when there is conflict

08_NOTES | Emerging Patterns
especially Pattern 6 and Pattern 7, because those patterns already constrain how much biomarker and frontier prestige can be allowed to dominate

Current Assessment

Current repository assessment:

importance to protocol integrity: foundational
importance to escalation control: foundational
importance to translational honesty: foundational
relevance to all protocol layers: system-wide

Open Questions

Which risk boundaries should be treated as absolute in all populations, and which should remain fit-dependent?
How much burden is acceptable before a support intervention stops being worth carrying?
How should unclear-but-not-yet-negative biomarker-function divergence be handled?
When does an exploratory intervention become too risky even to keep under active consideration?

Status

Foundational protocol-constraint file.

This file should be treated as the boundary layer that prevents protocol design from drifting past what the organism, the evidence, and the current level of interpretation can actually support.