AI Agent Governance — free for every LEAP 2026 attendee

AI Agent Governance by Dr. Fouad Bousetouane is free to download for anyone attending LEAP 2026 — 31 August to 3 September 2026 at the Riyadh Exhibition & Convention Centre, Malham, Riyadh, Saudi Arabia — in Arabic and in English. Enter the code handed out at the ProofAgent stand and both editions unlock on this page. No email address, no account, no newsletter.

One book, two editions

AI Agent Governance (English) is 157 printed pages across 17 chapters and 17,847 words: an end-to-end guide to evaluating, governing and releasing AI agents. حوكمة وكلاء الذكاء الاصطناعي (Arabic) is 188 printed pages across 17 chapters and 20,864 words, written in Arabic rather than machine translated, with its own Arabic vocabulary for authority, context, controls and release gates. Both are licensed CC BY 4.0 and archived on Zenodo under DOI 10.5281/zenodo.22037567.

Why these two books matter

Capability is not production readiness. An AI agent that demos well can still leak data on turn four, break a policy when flattered, or fire a tool it should have refused. The book separates what an agent can do from what it is allowed to do, and makes the second one testable — then turns the results into an evidence chain an auditor accepts rather than a dashboard screenshot.

What is inside

Seven parts, running from what an agent actually is to a signed release record: From AI Agents to Production Readiness; Understanding the AI Agent; Evaluation and Context; Compliance and Controls; Governance and Release; Applying the Framework; and two worked enterprise use cases governed with the same method against different bars.

About the author

Dr. Fouad Bousetouane is an award-winning AI expert, author and innovator, and the founder of ProofAgent.ai — the first platform for end-to-end AI agent evaluation and governance. He is an Adjunct Professor of AI and Agentic Systems at the University of Chicago, holds a Ph.D. in Artificial Intelligence, and was a postdoctoral scholar at the University of Nevada, Las Vegas.

He has led AI initiatives globally at Fortune 500 companies in the United States, including United Airlines and Grainger, and has collaborated on projects with AWS, Google and NVIDIA. He holds more than 50 patents and trade secrets in advanced AI systems, and authored the research ProofAgent is built on: the ProofAgent Harness, the Human-on-the-Bridge paradigm, The Context Fails First, and the PAI production-readiness index. He is a champion of responsible AI, and of the idea that an agent should earn its authority rather than be granted it.

Recognition: Top 30 AI Scientists (MIT Technology Review), the Timmy Award for Best Tech Manager in Chicago, Marquis Emerging Leaders, America's Most Innovative Teams (Fortune), and judge for the AWS Generative AI Awards.

Brought to you by

ProofAgent, the AI agent governance platform the book is written against; LEAP 2026 in Riyadh, which hosts the announcement and where the code is handed out; and DeepMinds, the DeepTech ventures studio bringing the Arabic edition to teams across the region.

Is AI Agent Governance free for LEAP 2026 attendees?

Yes. Both the English and the Arabic edition are free to download for anyone attending LEAP 2026 in Riyadh. Enter the code handed out at the ProofAgent stand and both PDFs unlock on this page. No email address, no account, no newsletter.

Where do I get the LEAP 2026 download code?

The code is given out at LEAP 2026 in Riyadh: at the ProofAgent stand, in the sessions, and on the QR poster. Scanning the QR code opens this page with the code already applied, so both editions unlock without typing anything.

Is the book available in Arabic?

Yes. The Arabic edition is a full book of 17 chapters and 188 printed pages, written in Arabic rather than machine translated, and it carries its own Arabic vocabulary for agent governance. It downloads from this page with the same LEAP 2026 code.

What does AI Agent Governance cover?

It covers one decision: whether an AI agent is ready for production. How to evaluate its behaviour under sustained adversarial pressure, how to score the quality of its context, how to turn a regulatory requirement into a control you can actually test, and how to record a release decision that holds up in front of an auditor.

Who wrote AI Agent Governance?

Dr. Fouad Bousetouane, founder of ProofAgent and author of five published papers on AI agent evaluation and governance. The book is the applied companion to that research: the readiness index, the release gate and the AI Bill of Materials, written as a method a team can follow on Monday morning.

Can I cite or reuse the book?

Yes. It is archived on Zenodo under the permanent DOI 10.5281/zenodo.22037567 and released under CC BY 4.0, so it can be quoted, adapted and redistributed with attribution — including in university course material and internal governance documentation.

Do I need a ProofAgent account to download it?

No. The download is a PDF and needs nothing but the LEAP 2026 code. There is no sign-up, no trial, no credit card and no sales follow-up attached to it.