Independent Research

Epistemic Red Teaming

Evaluating frontier AI models for adversarial robustness, epistemic integrity, and complex failure modes through rigorous safety auditing.

Core Research Focus

Epistemic Auditing

Evaluating narrative smoothing, sycophancy, capability inflation, and grounded reasoning boundaries under high-complexity prompts.

Adversarial Stress Testing

Surfacing deep contextual degradation, state recovery failures, and frontier reasoning edge cases across agentic workflows.

Inquiries & Engagement

For research inquiries, audit collaboration, or institutional outreach: