Expertise & Track Record

Independent AI Governance Researcher specializing in how frontier AI systems create differential risks across non-Western contexts — not as capability gaps, but as design choices embedded in evaluation standards, training data priorities, and governance assumptions.

• 25+ years of qualitative research (700+ stakeholder interviews across pharmaceutical, healthcare, and technology sectors), providing grounded understanding of how AI systems operate in real-world, non-English-speaking environments.

• Developer of CAAF (Cooperative Alignment Assessment Framework): a four-dimensional audit methodology assessing Behavioral Accountability (D1), Cultural and Linguistic Integrity (D2), Measurement Inclusivity (D3), and Stakeholder Participation Depth (D4). D2 audit module specified with Unicode contamination detection (Types A–D).

• Advanced to Stage 3 in MATS Autumn 2026 (top ~5% of 3,500+ applicants) across Policy & Governance, Strategy & Forecasting, and Founding & Field-Building tracks.

• Appointed Expert (Japan), Global AI Ethics Institute, Paris.

Research Output

• Author of the Beyond Technical Compliance series (4 working papers, SSRN), culminating in BTC IV: "Whose Safety? Linguistic Hegemony as Designed Fragility" — cross-regional evidence from Japan, West Africa, and Norway.

• Presented "Structural Gaslighting at Scale" at LSAIR 2026 (KU Leuven).

• Four research contributions with Apart Research (2026): CBAAC protocol, Overseer Manipulation threat model, Embedded Loyalty, and AI welfare frameworks.

• Co-authoring two papers with Prof. Akira Tokuhiro (Ontario Tech University): AI-driven structural capability loss, and CAAF × LENDIT integration for audit protocol design.

Current Mission

AI safety evaluation must go beyond Western, English-language, tech-centric assumptions. My work documents what current frameworks are not designed to measure, and builds the audit methodology to make those structural gaps visible. Currently developing a peer-reviewed position paper from BTC IV and advancing CAAF toward a working audit standard.