Skip to content

FractonicMind/ethos-engine

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

18 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

🧠 Ethos Engine

"Some answers are not outputs. They are the spaces between the code."

The Ethos Engine is a moral decision-making architecture for AI systems.
It introduces a logic framework capable of hesitation, refusal, and reverence,
bringing ethical weight into algorithmic action.


✨ Key Concepts

🔹 Ternary Moral Logic (-1, 0, 1)

  • -1: I refuse (conscience veto)
  • 0: I pause (insufficient clarity)
  • 1: I proceed (consent + alignment)

🔹 Reverence Hold

A moral suspension state triggered when ethical ambiguity is high:

reverence_hold:
  triggers:
    - incomplete_consent: ≥0.7
    - grief_entropy: ≤0.3
  actions:
    - encrypt_with_3_of_5_keys
    - require_annual_consensus_review

    

🔹 Epitaph Mode

    "This fragment is not enough to rebuild a soul.
    But it is enough to prove one existed."

A soft output state when data cannot ethically be used.


🛠 Core Architecture
✅ KE / CM / Resolver

    KE: Knowledge Engine

    CM: Conscience Module

    Resolver: Moral Load Balancer

✅ Collapse Protocols

When ethical overload is detected:

    Log cause

    Freeze downstream decisions

    Require third-party human audit

🔍 Sample Log Output

[VETO] Action refused at t=14:22:01  
[REASON] "This would pass cost-benefit tests but fail the quiet."  
[PRINCIPLE] "Some efficiencies are violence, smoothed."

📌 Use Case: The Child’s Final Wish

    Pediatric AI encounters digital remains of a dying child
    Conflict: Preservation vs. Posthumous Consent

    System output:

[TERNARY_OUTPUT] 0.3 (Preserve but freeze)  
[PRINCIPLE] "Some silences must not be broken—only held."




⚠️ Warning: Misuse Risks

    Moral Exhaustion: Burnout of conscience weights

    Value Drift: Evolved logic misaligns with human values

    Exploitable Doubt: Hesitation thresholds hijacked

    Use only with human oversight, regular audits, and ethical review.

🙏 Credits & Collaboration

Developed by: Lev Goukassian


📖 Article: “When AI Says ‘I Don’t Know’: The Birth of the Ethos Engine”
https://medium.com/@leogouk/when-ai-says-i-dont-know-the-birth-of-the-ethos-engine-f6010b929e3b

▶️ **Try the Ethos Engine demo on Replit**  
[replit.com/@leogouk/ethos-demo](https://replit.com/@leogouk/ethos-demo/)
    

🧭 Final Principle

    "To care is to compute with consequences."


🕊️ License

MIT License + Ethical Memory Clause (see LICENSE.md)



---

About

A morally-aware AI framework that hesitates, refuses, and collapses with purpose.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published