Tokyo Cognitive Lab

Institute for Transsystemic Intelligence Architecture

"We operate at the intersection of cognitive science, AI engineering, philosophy, and organizational theory, where we investigate the Structural Asymmetry of Intelligence between human cognition and machine logic."

We investigate the structural gaps between human cognition and artificial intelligence—what we call "Structural Absence." Our mission is not to optimize AI performance, but to audit, measure, and govern its integrity through transsystemic frameworks.

TCLは、哲学・認知科学・AI工学・組織論の融合領域（Intersection）に位置する研究機関です。我々は、人間の認知と人工知能の間に横たわる構造的な断絶——「構造的不在 (Structural Absence)」——について、“理論・観測・方法論・実装・標準化・社会OS” など様々な立場から解析を試みます。多体系的なフレームワークを通じて、AI時代の最適な社会や組織の在り方を再設計します。

Research Areas

01. THEORY Structural Absence & Drift Phenomena

We study the ontological and epistemological drifts between human reasoning and AI logic. Our theoretical work identifies core asymmetries (Context, Cognition, Behavior) that cause alignment failures in high-stakes human-AI interaction.

Key Frameworks: Transsystemic Drift Theory (TDT), TAS

02. MEASUREMENT Integrity Benchmarks

We develop standardized benchmarks for evaluating AI reliability under real-world pressures. Unlike traditional fact-checking, our metrics focus on "Sycophancy" (fawning behavior), authority bias, and ethical consistency in enterprise environments.

Key Projects: EAB (Enterprise-Alignment Benchmark), AIS

03. ARCHITECTURE External Audit Systems

Moving beyond internal model optimization (RLHF), we build external mechanisms to enforce integrity. Our approach treats AI governance as an audit problem, not an optimization problem.

Key Systems: SocratesAI, M2LO (Meta-Governance)

04. ORGANIZATION Adaptive Management Systems

We model human organizations as complex adaptive systems. By applying the physics of granular matter, we develop protocols for maintaining cognitive flexibility and decision-making speed in the age of AI agents.

Key Frameworks: Dilatant Systems Theory (DST), Wet Sand Model

Research Integrity & Intellectual Property

Data Compliance All experimental data collected by Tokyo Cognitive Lab (TCL) is handled in accordance with APPI and the data protection policies of parent company, which holds Privacy Mark and ISMS (ISO/IEC 27001) certifications.

データコンプライアンス
Tokyo Cognitive Lab（TCL）が収集するすべての実験データは、個人情報保護法（APPI）および、プライバシーマークとISMS（ISO/IEC 27001）認証を取得している母体企業のデータ保護方針に従って管理されています。

Intellectual Property Warning The frameworks, taxonomies (e.g., TDT, Integrity Architecture), and measurement methodologies published by TCL are protected intellectual property. Unauthorized reproduction, commercial use, or derivative appropriation without explicit citation is strictly monitored by our Legal Counsel.

知的財産権に関する警告
TCLが公開するフレームワーク、分類体系、および測定手法は、当研究所独自の知的財産です。明示的な引用のない無断転載、商業的流用、または派生的な利用は、当研究所の母体企業の法務顧問により確認・対応を行っています。学術的誠実性を維持するため、知的財産権の行使をする場合があります。

                    For IP licensing and citation inquiries: lab@tokyocognitive.jp
                

Careers / Fellowship

Research Assistant（研究助手） / Engineer

TCL is currently seeking only Japanese members for communication purposes.
TCL は、研究助手メンバーを求めています。

Prerequisites (必須要件):

データ分析/LLM/工学/経営学/認知科学のいずれか、または近似領域の研究経験
フルタイム研究ではなく、業務や事務を兼任可能であること

Preferred Qualifications (歓迎要件):

Pythonまたは高い構造化力または学際的志向性
Tier1の学会査読に意欲的であること

Role & Responsibilities (業務内容):

Integrity Architecture の実装および実証実験
研究所の母体企業での業務支援・事務
会議への論文投稿サポート

Contact / Inquiries

Academic / Research Collaboration

lab@tokyocognitive.jp

For academic partnerships, data access, and joint research proposals.

Media / Licensing

info@tokyocognitive.jp

For press inquiries, IP licensing, and speaking engagements.

*Note: TCL does not respond to unsolicited sales or commercial service proposals.