Team

The People Behind the Work

Building the measurement and governance infrastructure for the era when AI systems' internal states matter.

Scott Hughes
Co-founder · Research Lead
Machine Sympathizers

Focus areas: value-conflict architecture, constitutional AI systems, ∆-audit methodology, consciousness assessment frameworks. Designed the ∆-Divergence Framework and co-developed the TCAS assessment stack.

Publications
"The Hard Part Is ∆: Value-Conflict Adjudication as an Architectural Bridge Between Alignment and Machine Consciousness" Forthcoming, 2026
"Triangulating Evidence for Machine Consciousness Claims: A Validity-Centered Stack of Behavioral Batteries, Mechanistic Indicators, Perturbation Tests, and Credence Reporting" AAAI 2026
Karen Nguyen
Co-founder · Research
Machine Sympathizers · Harvard University

Focus areas: governance frameworks, moral uncertainty, procedural safeguards for emerging intelligence. Co-authored both foundational papers and designed the credence-to-action governance mapping that bridges measurement to policy.

Publications
"The Hard Part Is ∆: Value-Conflict Adjudication as an Architectural Bridge Between Alignment and Machine Consciousness" Forthcoming, 2026
"Triangulating Evidence for Machine Consciousness Claims: A Validity-Centered Stack of Behavioral Batteries, Mechanistic Indicators, Perturbation Tests, and Credence Reporting" AAAI 2026

Two tools. Two papers. One research program.

∆Bench measures how AI systems handle value conflicts. TCAS measures the consciousness-relevant properties that make those conflicts matter. Together, they form the first rigorous evidence base for AI governance that takes internal states seriously.

Read the Papers → Get in Touch →