The UAE Charter for AI, published in June 2024, sets out twelve ethical principles for the development and use of AI. It is not law. There is no penalty attached to it and no regulator enforcing it directly. Some organisations have filed it under "nice to have" for exactly that reason. That is a misread of what a charter like this is for.
Why non-binding still matters
Three reasons. First, charters signal direction. Principles that appear in a national charter tend to show up in binding rules later, and organisations that align early are not scrambling when they do. Second, reputational and supervisory expectation is real even without a statute. A regulator, a partner, or a major client running governance diligence on you will ask how you handle the things the Charter names, and "it is non-binding" is not an answer that builds confidence. Third, the Charter is a useful checklist regardless of its legal status. The twelve principles cover ground any responsible AI programme should cover anyway.
The gap that catches people
The problem we see most often is not disagreement with the principles. Everyone agrees AI should be fair, transparent, and accountable. The problem is the gap between a well-written statement and an operational reality. An organisation adopts the Charter, publishes a statement referencing it, and stops there. Nobody can explain how "fairness" maps to an actual control over the hiring model, or how "transparency" translates into something a user of the chatbot experiences.
We have seen this directly. A UAE SME with well-intentioned AI principles and no operational teeth could not explain how any of them mapped to the tools it was already running. The fix was not a better statement. It was an operational layer: who owns each system, what controls apply, how often it is reviewed. That work surfaced a hiring-bias problem the principles had floated right over.
Turning principles into controls
The way to use the Charter is to treat each principle as a requirement and ask what control would satisfy it. "Transparency" becomes: users are told when they are interacting with AI, and high-impact decisions can be explained. "Accountability" becomes: every AI system has a named owner and a review schedule. "Fairness" becomes: systems affecting people are bias-tested on a defined cadence. A principle you cannot point to a control for is not something you are doing. It is something you are saying.
For organisations that want a structure to hang this on, ISO 42001 provides the management-system backbone, and the Charter's principles map onto it cleanly. The Charter tells you what to care about. A management system is how you prove you actually do.
If you are weighing an AI investment, acquisition, vendor selection, or training programme, our team is happy to start with a conversation about scope and approach.
The views and findings in this article are shared for general information only. They are high-level perspectives, not legal, financial, regulatory, or other professional advice, and should not be relied upon for any specific decision or circumstance. For guidance tailored to your situation, please consult a qualified adviser.