RESEARCH
Updated August 2026
The research map
The frameworks here come from enterprise AI governance work in regulated environments. Each one addresses a gap that became visible during real deployment, not in theory. An agent can be configured, logged, and reviewed, and still have no written record showing who authorized it, what it was supposed to accomplish, or who is accountable when it produces a harmful output. This research addresses that gap directly. Every framework and every brief here is built from named primary sources. No product agenda. No consulting pitch.
Sougata Roy | sougataroy.com | Views are my own
Scope
Seven areas
Publication record
Versioned and citable
Framework library
10 operational outputs
Research posture
Primary-source led
Use this first
Research is easier to use when the first click matches the work in front of you. Choose the path that fits today's question, then move into the supporting evidence.
Need an artifact
Use a named, numbered object when the work is a review, approval, workshop, or examiner conversation.
Open the framework libraryNeed the argument
Start with the white paper when the question is why technical controls need an organizational record above them.
Read Who Owns the Agent?Need the current signal
Use the newsletter for current governance movement, primary sources, and implications for regulated environments.
Read The Governance Gap01 / Research areas
Nine entry points into the governance system, from agent accountability and regulation to the technical control plane and examiner readiness.
Who is responsible when an AI agent makes a wrong decision at scale.
Open research
EU AI Act, NIST AI RMF, and sector requirements on Microsoft 365.
Open research
Classifying and monitoring sensitive data before AI can reach it.
Open research
Governance architecture before agents proliferate across your environment.
Open research
How Microsoft is becoming the control plane for enterprise AI governance.
Open research
What an examiner will ask when your AI governance posture is tested.
Open research
A concise view of whether controls, ownership, and oversight are keeping pace.
Open research
The citable framework library for enterprise AI agent authorization records.
Open research
The white paper on accountability architecture at the individual agent level.
Open research
02 / White papers
Read the argument, cite the source, or take the document into a meeting. Each publication has a stable version, date, and download path.
Framework Library
The Enterprise AI Agent Governance Framework Library
A citable v1.0 framework library for CISOs, CTOs, security architects, and board members. Five core concepts, nine operating frameworks, and one governance record are compiled for reading, printing, and internal circulation.
v1.0 | July 7, 2026 | DOI 10.5281/zenodo.21245690 | CC BY 4.0
Editable copyWhite Paper
The Intent Architecture Stack
A framework white paper for CISOs, CTOs, security architects, and board members. The Intent Architecture Stack gives the accountability requirement in NIST AI RMF, EU AI Act, and the CSA Agentic Profile its organizational form at the individual agent level, before deployment, not after an incident.
v1.0 | May 2026 | DOI 10.5281/zenodo.20481551 | CC BY 4.0
Editable copy03 / Frameworks
Ten operational frameworks built from enterprise AI governance work in regulated environments. Each one produces a governance artifact an examiner can inspect.