One of the things I hear with some frequency when I raise AI ethics and governance with institutional leaders across Lebanon and the Arab region is a version of the following response. That is a concern for regulators in the United States and Europe. We are focused on adoption first. Ethics can wait.
I want to challenge that framing directly, because I think it rests on a misunderstanding of what AI ethics actually is and why it matters for Arab institutions specifically, not despite our context but because of it.
AI ethics is not a constraint on Arab region development. It is a condition for the kind of trust that makes development sustainable.
Why do some Arab leaders see AI ethics as a Western concern?
The perception that AI ethics is a Western export, layered onto technology developed elsewhere and governed by regulatory cultures that do not map onto our region, is understandable. Much of the public discourse on AI ethics has emerged from American and European academic and policy contexts. The frameworks produced by major Western technology companies do reflect assumptions and priorities that do not always translate directly to the Arab region’s institutional landscape, whether in Lebanon, the Gulf, Egypt, or the Levant.
But the underlying problems that AI ethics is trying to address are not Western problems. Algorithmic bias, the concentration of decision making power inside unauditable systems, the erosion of individual privacy through behavioral data collection, the manipulation of information environments at scale: none of these risks is geographically confined.
And several of them have a particular resonance in the Arab region, where trust in institutions has been shaped by experiences that differ from those of Western populations, where regulatory capacity is developing rapidly but unevenly, and where the pace of AI adoption is accelerating faster than the governance frameworks are being built.
What is the cost of treating ethics as optional?
The cost of treating AI ethics as optional or deferrable is not primarily a reputational cost, although that matters too. It is an operational and strategic cost that compounds over time.
An institution that deploys AI without ethical boundaries does not know, with any confidence, what its AI systems are actually optimizing for. It may believe they are optimizing for the outcomes it has specified, but without the governance structures to audit and validate that, it is operating on assumption. And in regulated sectors like banking, insurance, and healthcare, operating on assumption in decision making processes is a risk that boards have a duty to manage, not an acceptable baseline.
More concretely, an institution that deploys AI without explainability and fairness requirements baked into procurement and deployment will eventually face a situation where an AI output produces harm, discrimination, or a regulatory violation that it cannot account for. When that happens in a context where trust is already fragile, the damage is not proportionate to the incident. It is disproportionate, because it confirms a fear that the institution was not managing its technology responsibly.
What would Arab region AI governance look like if we built it for our context?
This is the question I find most interesting, and most underexplored in the conversations I am having with institutions across Lebanon, the Gulf, Egypt, and the Levant.
A governance framework built for the Arab region’s institutional context would take seriously the specific trust dynamics between institutions and the populations they serve. It would account for the uneven regulatory landscape across the region and build in flexibility for that landscape to evolve. It would incorporate cultural and linguistic considerations into fairness requirements, recognizing that algorithmic bias can be embedded in training data that reflects social realities very different from those of the Arab region.
It would treat transparency as an active practice rather than a disclosure requirement. Institutions would not simply publish terms of service that describe their AI use. They would build communication practices that help customers understand, in plain language appropriate to their context, how AI is being used in decisions that affect them.
And it would position Arab region institutions as leaders in this space rather than followers. There is a genuine opportunity for institutions in this region to develop governance frameworks that address our specific context with real rigor, and to demonstrate that AI ethics is not a constraint on development but a condition for the kind of trust that makes development sustainable.
I work on exactly these questions as an AI expert covering Lebanon and the Arab region. If this essay raises questions relevant to your institution’s AI strategy, you can find more of my thinking at hildamaaloufmelki.com, and in my book AI Simplified at hildamaaloufmelki.com/signature-book.
About the Author
Hilda Maalouf Melki is an Oxford-Certified AI Expert, an Official Member of the Forbes Business Development Council, Chair of the AI & Innovation Committee at Lions Clubs International – District 351 (Lebanon, Jordan & Palestine), and the author of AI Simplified (الذكاء الاصطناعي ببساطة).
With more than 25 years of experience spanning banking, digital transformation, brand strategy, and artificial intelligence, she advises executives and institutions on AI strategy, AI governance, responsible AI adoption, and institutional transformation. Through her writing, speaking, and advisory work, she helps leaders move beyond AI hype to practical, strategic implementation that creates measurable business value.
Follow AI Simplified on Substack for evidence-based insights on artificial intelligence, leadership, governance, and the future of business.
Website: hildamaaloufmelki.com