TRACE AI Labs is the joint research of Mika Okamoto and Ansel Erol on making AI agents trustworthy, auditable, and efficient. We care about leaving a clear trace — the audit trails that make agent behavior explainable and verifiable.
- AI agent compliance & enterprise AI — whether LLM agents actually follow the rules they're given, and why they don't.
- LLM & agent routing — explainable, cost-aware routing that picks the right model for each task and shows its reasoning.
- Efficient AI — frontier-level results under real cost and latency budgets.
…and adjacent directions in agentic evaluation and explainability as they come up.
- PACT: Can Enterprise AI Assistants Be Trusted Under Pressure? — an LLM compliance and AI safety benchmark of 3,364 items pairing a company rule with a convenient shortcut that breaks it, across 48 realistic scenarios, 12 regulated domains, and nine psychology-grounded pressures. Ordinary pressure raises violation rates by 65% across 22 models, and none clears the bar for unsupervised use. Preprint · Interactive results · Code · Dataset — under review (2026).
- Why Do AI Agents Break Rules? How Framing, Context, and Social Signals Shape Compliance — how rule framing, enforcement, and workplace pressure shape regulatory compliance across twelve LLM agents. Paper · Interactive results · Code & data — AAAI/ACM Conference on AI, Ethics, and Society (2026), Conference on Language Modeling Workshop on Agent Behavior (2026).
- Explainable Model Routing for Agentic Workflows — interpretable, auditable model routing for agentic systems. Paper — ACM CHI Workshop on Human-Centered Explainable AI (2026).
- Trust by Design: Skill Profiles for Transparent, Cost-Aware LLM Routing — skill-decomposed, budget-aware model selection. Paper — MLSys Young Professionals Symposium (2025).