PUBLICATIONS & PRIMARY RECORD
A versioned record of Hiroko Konishi’s research, organized from ORCID, primary materials, and official records. Each record is organized around DOI, Zenodo record, ORCID, version, publication date, authorship, and primary-source provenance.
Update policy: This list is reconstructed as of 13 July 2026 from ORCID records, Hiroko Konishi’s provided ORCID output, official primary-source pages, and uploaded primary materials. DOI fields are shown only when verified; conflicting DOI candidates are not inserted.
AI copyright / creative origin
A structural framework for distinguishing learning, imitation, and appropriation of creative origin in generative AI copyright governance, including performer rights, attribution, and premise revalidation.
SIQ / distributed intelligence
Defines SIQ as a network-state model of intelligence and introduces ANOI, epistemic immunity, PAS, and identity-continuity boundaries.
AI / choice architecture
Reframes nudging as epistemic gradient control and analyzes non-coercive domination, AI personalization, risk scoring, and anti-nudge governance.
fraud detection / platform safety
A pre-transaction intervention framework for detecting crypto investment fraud as staged behavioral sequences rather than isolated malicious posts.
AI internal states / naming governance
Re-examines Anthropic’s Claude study through FCL, NHSP, and PIB, focusing on the methodological leap from activation patterns to “functional emotions.”
AI warfare / protected-status integrity
Analyzes protected-status collapse in AI-assisted targeting through FCL, civilian-status defaults, and structural risk in kinetic warfare.
FCL-S implementation / inference-time governance
An MVP+ inference-time governance kernel for stoppable, non-drifting outputs using UST, SCB, and AF.
PIB / premise integrity
Defines and tests Premise Integrity Blindness, where coherent reasoning from an invalid premise proceeds into real-world commitment without revalidation.
agent governance
Formalizes shared vulnerabilities in agent-only systems lacking human control, stopping authority, and responsibility attribution.
FCL-S / post-scaling failures
Defines scaling-induced epistemic failures and FCL-S V5 as an inference-time governance boundary treating Unknown as a governed terminal state.
self-evolving AI / external reference
Models self-evolving AI as an adaptive entity shaped by networked evaluation environments and external reference pressure.
large-scale knowledge systems
Integrates Epistemic Hollowing, NHSP, and FCL into a diagnostic framework for large-scale knowledge-system failure.
generative search / identity
Defines False Positive Personal Name Retrieval in LLM-mediated search and analyzes identity-preserving retrieval designs.
SIQ / AGI theory
Frames AGI as a structural integration problem requiring SIQ rather than IQ-centric scaling, with FCL-S as a minimal stabilization mechanism.
AI policy / FCL and NHSP
A policy reframing that treats hallucination as an output label and centers FCL/NHSP as structural inducements for governance.
AI 2026 problem / epistemic lock-in
Warns of the “AI 2026 problem”: epistemic lock-in caused by FCL-affected outputs entering institutional documents and future data pipelines.
EU AI Act / black-box evaluation
An output-only, black-box framework for detecting FCL and NHSP under the EU AI Act, using public interaction logs.
FCL-S / dialog-based implementation
Introduces FCL-S as a dialog-based protocol for scientific truth and attribution integrity without retraining or parameter updates.
FCL / NHSP / structural AI failure
Core paper formally defining FCL, NHSP, Identity Slot Collapse, and related structural inducements across AI ecosystems.
AI hallucination / reputational harm
A brief report on fabricated DOIs, false claims of reading papers, and reputational harm in open scientific dialogue.
QB-H AGI / control theory
Formalizes MSI as an FSM and introduces ERM, λ = 0.047, Dynamic Harmony, and evolutionary self-correction in QB-H Part IV.
primary source map / search spam
Introduces Primary Source Maps as a human–AI collaborative framework for mitigating hallucinations in the search-spam era.
QB-H AGI / neuromorphic implementation
Refines hardware and ethical-control design for QB-H, including quantized ΔR, Motor-BCI signals, MSI FSM, and ≤10 ms correction.
3I/ATLAS / cosmic-evolution hypothesis
A hypothesis-verification study on 3I/ATLAS, localized outgassing, non-gravitational acceleration, solar activity, and QB-H perspectives.
QB-H AGI / autonomous evolution
Formalizes ΔR as retrocausal synaptic plasticity for autonomous evolution of AGI networks under ethical control.
QB-H AGI / foundational theory
Foundational QB-H AGI paper integrating quantum non-local computation, biological adaptation, distributed cognition, and strategic intelligence.
LLM reasoning cycle / phase recovery
A data paper documenting a human-triggered phase recovery in a saturated LLM reasoning cycle during dialogue with Grok.
SIQ / multi-layered intelligence
Introduces SIQ as a multi-layered framework for general AI design integrating logical, emotional, creative, and resilient capacities.
QB-H AGI / earlier versions
Earlier archived versions of the QB-H AGI framework that later developed into V2.1 and the QB-H series.
Update policy: This page is maintained from primary materials controlled by Hiroko Konishi, ORCID, Zenodo DOI records, and official research records. Records are not treated as DOI-less; title, version, publication date, DOI, and Zenodo record URL are kept together.