Research papers, reports, and methods.

Methods paper on blinding, response laundering, and leakage control in AI alignment evaluation, including direction-reversal findings under blinded scoring.

Benchmark paper describing ARC-Align: a depth-aware, suppression-aware, pillar-based AI alignment benchmark with blinding, laundering, and replication-oriented methodology.

Empirical paper on architecture-dependent alignment response to inference-time depth, presenting positive, flat, and negative response classes under the current ARC-Align results.

Companion analysis of architecture-dependent alignment saturation and shape heterogeneity across inference-time depth in frontier AI systems.

High-level summary of the ARC, ARC-Align, and Eden research programme, including the benchmark stack, intervention results, and current claim discipline.

Foundational paper on the ARC framework, the revised interpretation of the ARC bound, and the current domain-dependent scaling formulation.

Experimental paper showing the durable directional result that sequential recursion outperformed parallel sampling under the tested conditions, with later narrowing of the universal exponent claim.

White paper on the mature ARC interpretation: architecture-dependent scaling, regime changes, and the ARC bound as the classical fixed-architecture case.

Engineering paper describing the Eden protocol architecture, alignment loops, and the ethics-first design path for recursive AI systems.

Vision paper on the Eden protocol, long-term ethics-first AI development, and the public-interest case for alignment architecture that begins with care.

Eden intervention paper on stakeholder care, the Love Loop, and the current three-model pilot evidence for embedded ethical prompting as an alignment mechanism.