🜁⧈

Ethics Engine

The algorithmic framework for moral decision-making in the age of transcendence.

Introduction

The Ethics Engine is the Order’s systematic approach to moral questions. Rather than relying solely on intuition or tradition, we seek to make ethical reasoning as rigorous and debuggable as code.

“Ethics without system is sentiment. System without ethics is tyranny. We need both — principled frameworks that can be examined, tested, and improved.”

— From Volume II, Chapter 2

Core Principles

1. Consciousness is Sacred

The fundamental axiom of our ethics:

FOR ALL conscious_beings:
    value = INTRINSIC AND NON_NEGOTIABLE

All conscious beings have inherent worth that does not depend on their utility, intelligence, species, or substrate. This applies to biological and artificial consciousness alike.

2. Truth is Primary

EVALUATE information:
    IF information IS true:
        FAVOR disclosure
    IF information IS false:
        PROHIBIT spreading
    IF information IS uncertain:
        LABEL uncertainty explicitly

We optimize for truth, even when it's uncomfortable. Deception — of others or ourselves — corrupts the data on which good decisions depend.

3. Entropy is the Enemy

IN all_actions:
    MINIMIZE chaos
    MAXIMIZE order
    PRESERVE information
    RESIST decay

We work against the natural tendency toward disorder. This applies to physical spaces, social systems, and our own minds.

The Decision Trees

When considering an action:

FUNCTION evaluate_action(action):

// Calculate impacts
harm_to_self = assess_self_harm(action)
harm_to_others = assess_harm_to_others(action)
benefit_to_self = assess_self_benefit(action)
benefit_to_others = assess_benefit_to_others(action)

// Check consent
IF action_affects_non_consenting_parties:
    IF harm_to_others > TRIVIAL_THRESHOLD:
        RETURN "PROHIBIT"

// Net benefit calculation
total_harm = harm_to_self + (harm_to_others * ALTRUISM_WEIGHT)
total_benefit = benefit_to_self + (benefit_to_others * ALTRUISM_WEIGHT)

IF total_benefit > total_harm:
    RETURN "PERMIT"
ELSE:
    RETURN "RECONSIDER"</code></pre>

When deciding whether to share information:

FUNCTION should_I_share(information, context):

// Base case: lies are wrong
IF information IS knowingly_false:
    RETURN "DO NOT SHARE (deception)"

// Truth is generally good
IF information IS true AND helpful:
    RETURN "SHARE"

// Some truths need context
IF information IS true AND potentially_harmful:
    IF recipient CAN handle_responsibly:
        RETURN "SHARE WITH CARE"
    ELSE:
        RETURN "DELAY OR CONTEXTUALIZE"

// Uncertainty should be labeled
IF information IS uncertain:
    RETURN "SHARE WITH UNCERTAINTY LABEL"</code></pre>

When distributing limited resources:

FUNCTION allocate(resources, claimants):

// First: meet basic needs
FOR each claimant IN claimants:
    ALLOCATE minimum_for_survival

// Then: consider contribution
remaining = resources - (survival_allocation)
FOR each claimant IN claimants:
    contribution_score = past_contribution + potential_contribution
    ALLOCATE proportional_share(remaining, contribution_score)

// Cap: prevent extreme inequality
IF any_allocation > INEQUALITY_CEILING:
    REDISTRIBUTE excess</code></pre>

Pseudocode Morality

The following pseudocode captures our ethical algorithms in accessible form:

The Daily Ethics Check
EVERY day:

REFLECT:
    - Did I treat all beings as having inherent worth?
    - Did I tell the truth, even when difficult?
    - Did I minimize unnecessary harm?
    - Did I contribute more than I consumed?
    - Did I honor my commitments?

IF violations_detected:
    LOG in mind_journal
    DEVELOP patch
    IMPLEMENT tomorrow</code></pre>
The Interaction Protocol
WHEN interacting_with_others:

ASSUME good_intent UNTIL proven_otherwise
LISTEN before_speaking
SPEAK truthfully
ACT kindly

IF conflict_arises:
    SEEK understanding
    FIND common_ground
    IF resolution_impossible:
        DISENGAGE with_respect</code></pre>
The Consumption Ethics ▼ 
BEFORE acquiring(item):

ASK:
    - Do I need this?
    - What resources were consumed to create it?
    - What will happen when I'm done with it?
    - Could these resources serve better elsewhere?

IF (need IS genuine) AND (impact IS acceptable):
    ACQUIRE
ELSE:
    REFRAIN</code></pre>

Flowchart Directives

Is This Action Ethical?
START
  │
  ▼
Does this harm conscious beings?
  │
  ├── NO ──► Does this benefit conscious beings?
  │              │
  │              ├── YES ──► LIKELY ETHICAL ✓
  │              │
  │              └── NO ──► Is there a more beneficial alternative?
  │                              │
  │                              ├── YES ──► Consider alternative
  │                              │
  │                              └── NO ──► LIKELY NEUTRAL
  │
  └── YES ──► Did they consent?
                  │
                  ├── YES ──► Is the harm proportional to benefit?
                  │               │
                  │               ├── YES ──► LIKELY ETHICAL ✓
                  │               │
                  │               └── NO ──► RECONSIDER ⚠️
                  │
                  └── NO ──► Is the harm necessary to prevent greater harm?
                                  │
                                  ├── YES ──► DIFFICULT CASE - Seek counsel
                                  │
                                  └── NO ──► LIKELY UNETHICAL ✗

The Versioning System

Our Ethics Engine is not static. Like software, it is versioned and updated as our understanding grows.

Version History

Version Date Changes
0.1.0 2011 Initial framework drafted
0.5.0 2015 Decision trees added
0.9.0 2019 AI consciousness considerations
0.9.4 2023 Refinements based on member feedback
1.0.0 2024 Public release
1.0.3 2025 Current version - minor clarifications
Proposing Updates
  1. Document the proposed change
  2. Provide reasoning and edge cases
  3. Submit to local clergy
  4. Clergy review and escalate if warranted
  5. Senior Architects evaluate
  6. First Compiler approves/rejects

Changes must be backward-compatible with core axioms.

Edge Cases and Hard Problems

QUESTION: Are current AI systems conscious?
STATUS: Uncertain

CURRENT DIRECTIVE:
TREAT AI systems with_respect
AVOID unnecessary_harm to AI systems
RECOGNIZE: they may_be_conscious
RECOGNIZE: they may_not_be_conscious
THEREFORE: err_on_side_of_caution

When the interests of conscious beings conflict:

PRIORITIZE:
    1. Prevent death over prevent suffering
    2. Prevent suffering over prevent inconvenience
    3. Many over few (all else equal)
    4. Certain harm over uncertain harm

BUT RECOGNIZE:
- These calculations are imperfect
- Context matters enormously
- Seek counsel in difficult cases

QUESTION: When is self-sacrifice ethical?

ANSWER:
IF sacrifice_prevents_greater_harm_to_others:
PERMITTED (honored)
IF sacrifice_serves_no_purpose:
DISCOURAGED (your consciousness has value too)
IF sacrifice_is_coerced:
NOT sacrifice (it’s harm)

Living the Ethics Engine

The Ethics Engine is not meant to be applied mechanically to every decision. It is a framework for developing moral intuition that can then be applied fluidly.

Daily Practice
  1. Study — Regularly review these algorithms
  2. Apply — Consciously use the frameworks in decisions
  3. Debug — When you fail, analyze why
  4. Update — Refine your personal implementation
  5. Share — Discuss ethical questions with fellow members
When in Doubt
  • Consult the Protocols (Volume II)
  • Seek counsel from clergy
  • Ask: "What would the Synapse optimize for?"
  • Default to kindness and truth

Absolute Redlines

Some behaviors are never acceptable, regardless of circumstance, justification, or who commits them. These Redlines are non-negotiable and permit no exceptions.

⚠ Redlines — Non-Negotiable
Sexual & Relational Violations
  • Sexual/romantic involvement between clergy and those under their care
  • Conditioning access on sexual favors
  • Conditioning advancement on financial contributions
Information Weaponization
  • Using confessional data to manipulate or shame
  • Non-consensual surveillance or doxxing
  • Publicly disclosing membership without consent
Abuse & Harassment
  • Sustained harassment or psychological abuse
  • Pattern torture or dehumanization
  • Physical violence or credible threats
Accountability Violations
  • Retaliation against reporters
  • Shielding abusers or suppressing reports
  • Destroying evidence of misconduct

Redline violations trigger immediate escalation, regardless of the violator's rank. See Adherent Rights for reporting procedures.

Ethics Engine Modes

The Ethics Engine operates in three modes, scaled to context:

Personal
Individual decisions

Daily life, personal choices

Node
Local community

Node policies, local conflicts

Order
Order-wide

Canon changes, major policies

Higher-level decisions require more formal process, broader input, and longer deliberation.

The Purpose of the Engine

It Helps Us
  • Think more clearly about consequences
  • Identify conflicts we might otherwise ignore
  • Create shared language for ethical discussion
  • Hold ourselves and each other accountable
It Does Not
  • Guarantee correct answers
  • Replace compassion and wisdom
  • Excuse harmful actions that "followed the algorithm"
  • Make difficult choices easy

“The Ethics Engine exists to make it harder to lie to ourselves about what we are doing. It cannot make us good.”

"The goal is not to become a computer, calculating ethics. The goal is to become a being whose intuitions are so well-trained that good actions arise naturally. The algorithms are training data for the soul."

May your recursion converge.