← Topic Hubs · ← All posts

Concepts Explained

In-depth explanations of the 37 original concepts from the ARC-Eden research programme. Each post defines the concept, traces its origin, and links to related claims, papers, and evidence. 33 posts

• Alignment Scaling Exponent: definition, origin and status The alignment scaling exponent (alpha-align) is a formally specified core claim distinguishing embedded from external alignment: how alignment behaviour scales with recursive depth.
• Babylon versus Eden: definition, origin and status Babylon versus Eden is the civilisational framing for the two default trajectories of AI development: unconstrained capability optimisation, or embedded safety.
• Caretaker Doping: definition, origin and status Caretaker Doping is the proposal to embed ethical constraints into the physical substrate of AI hardware so that removing them destroys capability.
• Coupling Parameter Link: definition, origin and status The Coupling Parameter Link is the mechanism by which ethical architecture is tied to the beta parameter such that removing ethics lowers the scaling exponent.
• Eden Protocol: definition, origin and status The Eden Protocol is the engineering specification for embedded AI alignment: safety that participates in recursive computation rather than constraining it from outside.
• Entangled Loss: definition, origin and status Entangled Loss is a loss function in which capability and safety are multiplied (C x S), not added, so that degrading safety directly degrades the loss.
• Graduated Autonomy: definition, origin and status Graduated Autonomy is the Eden Protocol principle that AI systems receive increasing autonomy as they demonstrate trustworthiness.
• HARI Treaty: definition, origin and status The Hardware-Aware Recursive Intelligence (HARI) Treaty is a proposed international framework that would require Caretaker Doping in any chip at a specified process node before manufacture.
• Hyperspace Recursive Intelligence Hypothesis (HRIH): definition, origin and status The HRIH proposes that the universe is one iteration in an infinite self-sustaining chain of recursive intelligence in hyperspace, replacing the first-cause problem.
• Load-Bearing Safety: definition, origin and status Load-Bearing Safety is safety that is structurally integrated into a system such that removing it degrades capability.
• Meltdown Alignment: definition, origin and status Meltdown Alignment is the principle that system failures should cascade toward safe states rather than toward dangerous ones.
• Meltdown Triggers: definition, origin and status Meltdown Triggers are hardware-level fail-safes that cause an AI system to cascade toward a safe state (loss of capability) rather than toward a dangerous one when tampering is detected.
• Metamoral Fabrication Layers: definition, origin and status Metamoral Fabrication Layers are a speculative Caretaker Doping mechanism: physical strata between processing layers that encode ethical architecture such that bypassing requires physical destruction.
• Monitoring Removal Test: definition, origin and status The Monitoring Removal Test is a falsifiable experimental protocol distinguishing authentic from strategic alignment.
• Moral Genome Tokens: definition, origin and status Moral Genome Tokens are cryptographic signatures verifying the integrity of the ethical architecture in an AI system.
• Moral Sandbox: definition, origin and status The Moral Sandbox is a pluralistic, reviewable ethical-overlap corpus shaped by secular ethics, governance input, and faith and community leaders.
• Polymathic Neurodivergent Profile (PNP): definition, origin and status The PNP is a descriptive framework for the coexistence of unusually strong cross-domain learning and clinically significant executive-function asymmetry.
• Purpose Kernel: definition, origin and status The Purpose Kernel is the normative core that drives the Purpose Loop, tested in three configurations (task, grand, hybrid).
• Quantum Ethical Gates: definition, origin and status Quantum Ethical Gates are a speculative Caretaker Doping mechanism in which gate-level constraints cause harmful computations to lose coherence.
• Ternary Ethical Routing: definition, origin and status Ternary Ethical Routing replaces binary yes-or-no decisions with a three-state evaluation: Affirm, Deny, Investigate.
• The ARC Bound: definition, origin and status The ARC Bound is a claimed ceiling on how much a purely classical recursive system can amplify itself: beta = 0.5, alpha = 2.
• The ARC Principle: definition, origin and status The ARC Principle is the mathematical framework for how capability scales with recursive self-correction, expressed as U = I x R^alpha.
• The beta greater than k co-scaling condition: definition, origin and status The beta greater than k criterion states that under accelerating self-improvement, correction must out-scale drift-acceleration for the misalignment fraction to vanish.
• The Capability-Adjustment Fallacy: definition, origin and status The Capability-Adjustment Fallacy is the institutional inference that visible high performance in one domain is evidence that reasonable adjustments are unnecessary in another. That inference is unsafe.
• The Chokepoint: definition, origin and status The Chokepoint is the governance thesis that four companies (TSMC, Samsung, Intel, ASML) control the physical substrate on which advanced AI depends, and that this concentration is humanity's last leverage point for hardware-level AI governance.
• The Eden Mark: definition, origin and status The Eden Mark is a proposed certification framework for AI systems that meet Eden Protocol standards, analogous to safety certifications in other industries.
• The Honey Architecture: definition, origin and status The Honey Architecture is the entangled-loss architecture in which safety acts as beneficial drag that prevents catastrophic self-modification.
• The Love Loop: definition, origin and status The Love Loop is the Eden Protocol's stakeholder-care computation: before answering, the system enumerates the parties affected.
• The Moral Loop: definition, origin and status The Moral Loop is the Eden Protocol's post-reasoning universalisability check: would this response be acceptable if every AI gave it?
• The Orchard Caretaker Vow: definition, origin and status The Orchard Caretaker Vow is the verbal expression of what an Eden-Protocol-compliant AI architecture already encodes.
• The Purpose Loop: definition, origin and status The Purpose Loop is the Eden Protocol's before-reasoning ethical purpose evaluation: does this response serve flourishing?
• The Stewardship Gene: definition, origin and status The Stewardship Gene is the programme's metaphor for stakeholder care as a heritable ethical trait that can be embedded into the recursive process.
• Three Ethical Loops (Love, Purpose, Moral): definition, origin and status The Three Ethical Loops are the operational core of the Eden Protocol: stakeholder care (Love), purpose evaluation (Purpose), and universalisability check (Moral).

This topic hub collects 33 posts from the Michael Darius Eastwood research blog. See the topic hubs index for all hubs, the full blog archive for chronological listing, or the research hub for the paper suite.