Accountability for AI agents

ProofAgent is the accountability layer for production AI agents, founded by Fouad Bousetouane. We turn agent risk into deployment evidence, from adversarial evaluation to gated release and compliance, built around the open-source ProofAgent Harness.

Our mission: governance for the responsible deployment of AI agents at scale

Capability is no longer the bottleneck. The risk that keeps AI agents out of production is behavior under pressure. Our mission is one continuous governance loop: evaluate how an agent behaves under adversarial multi-turn pressure across six metrics, gate every release on the result so a failing agent cannot ship, map each result to a compliance posture across the EU AI Act, NIST AI RMF, ISO/IEC 42001, and SOC 2, and keep agents assured in production with regression tracking and drift detection.

Our story

Production AI agents fail differently from models. They break three turns into a conversation, under pressure, through domain-specific failure modes a single answer test never sees. We built ProofAgent because the agent layer needs its own evaluation infrastructure, and because a launch decision should rest on evidence, not optimism. We started with the open-source Harness so any engineer could stress-test an agent locally, with their own model, and the governance platform scales that same engine into gated releases and compliance evidence across every agent a team runs.

Founder: Fouad Bousetouane, Ph.D.

Fouad Bousetouane, Ph.D., is a distinguished AI leader, innovator, and entrepreneur specializing in Generative AI, AI agents, enterprise AI, and AI governance. He is the Founder of ProofAgent, a Lecturer in Generative AI at the University of Chicago, and the author of the book AI Agents for Everyone. He holds a Ph.D. accredited by the University of Nevada, Las Vegas, has authored more than 50 AI publications, patents, and trade secrets, and has led AI transformation initiatives for Fortune 500 organizations. He is also the author of the research ProofAgent is built on: the ProofAgent Harness whitepaper (arXiv:2605.24134) and the Human-on-the-Bridge paradigm for scalable agent evaluation (arXiv:2606.16871). His work has been recognized with the Timmy Award for Best Tech Manager in Chicago, selection among the Top 30 AI Scientists by MIT Technology Review, the Marquis Emerging Leaders honor, and inclusion on Fortune's list of America's Most Innovative Teams, and he serves as a judge for the AWS Generative AI Awards.

ProofAgent — open-source AI agent evaluation ProofAgent Harness — open-source AI agent testing framework ProofAgent Harness documentation ProofAgent SDK documentation ProofAgent Platform — enterprise AI agent evaluation The 5-stage AI agent evaluation pipeline ProofAgent pricing — free open source to enterprise ProofAgent vs Phoenix, LangSmith, DeepEval, Langfuse Sample AI agent evaluation report Security and compliance for AI agent evaluation Open ecosystem for AI agent evaluation ProofAgent community blog About ProofAgent and founder Fouad Bousetouane Research behind ProofAgent — published papers ProofAgent Harness whitepaper Human-on-the-Bridge paper Privacy policy Terms of service ProofAgent Harness on GitHub proofagent-harness on PyPI ProofAgent Harness whitepaper (arXiv:2605.24134) Human-on-the-Bridge (arXiv:2606.16871)