National data regulators
Contributed to the definition of a national data index, the benchmark used to measure the data maturity of public entities.
Profile
The professional
I help GCC government and telecom organizations build trustworthy, governed data, the foundation enterprise AI depends on. I bring 30+ years across enterprise data, telecom, and technology to forge data into an asset you can trust: compliant, AI-ready, and built to drive value, whatever the mandate.
How I got here
I read Economics and Computer Science at the American University of Beirut, a dual major, and I have been working at the join between the two ever since. Economics taught me to ask what a thing is worth before asking how to build it. Computer science taught me how it actually gets built. Thirty years later that pairing is still the most useful thing I own, because the hard questions in enterprise AI are not technical questions. They are questions about value, evidence and trust.
The work itself has moved in one direction the whole time. I started in telecom and enterprise systems, where the data was large, messy and load bearing. That led to data warehousing and business intelligence, which led to enterprise architecture, which led to data governance, because every one of those disciplines eventually fails for the same reason: nobody can say what the data means, who owns it, or whether it can be trusted. Governance is what you build when you have been burned by that often enough.
AI governance is the same question asked again, one layer up. An organization that cannot describe its data cannot describe what its models learned, cannot explain a decision, and cannot show a regulator its work. So I did not move from data governance to AI governance. The ground moved, and the discipline came with it.
Two frameworks came out of that. The Two Wings AI Framework asks the two questions an AI program has to answer at once, whether it is the right AI to build and whether the AI is built right. The AI Readiness Diagnostic Framework measures, over nine dimensions, whether the data underneath is ready to carry any of it. Both are published, and the first is a formal paper.
Where this has been applied
Engagements are described by category rather than by name.
Public and private sectors
Data strategy, establishing the data governance office, operating models, policies and processes, and maturity uplift.
Telecom operators
Data governance, enterprise architecture, data warehousing and business intelligence, across the GCC, EMEA and the United States.
Since 2017
My work on AI is not recent. I established the brand in 2017 and was writing about the economics of AI in 2020, when the subject was a good deal less crowded than it is now. Those posts are still online, dates intact, at aiconomica.com and ai2022.com. I leave them up on purpose. Anyone can claim a decade of interest in a subject; it is more useful to be able to point at the receipts.
Practice
I work through Green Data, and on AI engagements under the Aiconomica brand. I hold an independent assurer position on the systems I assess. Independence here means conflict free: a builder cannot credibly audit its own work.
Languages: English (fluent), French (fluent), Arabic (fluent), Spanish (intermediate). Regions: GCC and EMEA.