UAE Central Bank AI rules banks 2026
The UAE Central Bank Just Gave Banks New AI Rules. Here Is What It Means for Your Money.
On February 23, 2026, the Central Bank of the UAE issued a formal guidance note on how banks and other licensed financial institutions must handle AI and machine learning. Banks already use AI constantly, approving loans, spotting fraud, running customer service chatbots. This guidance does not stop any of that. It sets rules for how it has to be done.
| VERDICT: A real, current regulatory framework, technically non-binding, but treated as active supervisory expectation from day one. The guidance applies to every licensed financial institution under CBUAE supervision, including banks and insurers. It centers on the concept of a high-impact decision, any AI-driven determination that materially affects your access to a financial product or service, like a loan approval, credit limit, or insurance pricing. For these decisions, banks must maintain governance and accountability at the board level, ensure fairness and non-discrimination, provide transparency about when AI was involved, and give customers the ability to request human review or an explanation. Though not legally binding in the strictest sense, the CBUAE has made clear it will treat this guidance as part of active supervisory review going forward. |
The Core Idea: High-Impact Decisions
The guidance’s central concept is the high-impact decision. That’s any AI-driven determination that materially affects your access to financial products or services. A loan rejection. A lower-than-expected credit limit. A higher insurance premium. A fraud flag on your account. If AI played a meaningful role in producing that outcome, it falls inside this category. And it carries specific obligations for the bank behind it.
What Banks Now Actually Have to Do
Governance and accountability sit at the top of the list. Boards and senior management are now explicitly responsible for how AI systems are chosen, monitored, and managed. Not just the technical teams building them. Fairness and non-discrimination is a second core principle. AI systems must not produce biased or discriminatory outcomes, particularly in loans, pricing, or insurance. These are areas where algorithmic bias has caused real, documented harm in other markets.
Transparency is the piece most directly relevant to you as a customer. Banks must maintain transparency with customers about when and how AI is being used in decisions that affect them. And critically, you must be able to request a human review or a clearer explanation of an AI-generated decision. Alternative arrangements must be available if you do not want to be subject to an automated decision at all.
The Third-Party Piece Worth Knowing
A significant share of the AI tools banks actually deploy come from outside vendors, not built in-house. The guidance closes an obvious loophole here directly. Institutions remain fully responsible for outsourced AI functions. They must be able to suspend the use of a third-party AI system immediately if risks or failures emerge. A bank cannot point to a vendor and disclaim responsibility for how that vendor’s AI treated your loan application.
Why This Landed Now
The timing is not accidental. A DFSA survey found generative AI usage among UAE financial institutions surged 166% between 2024 and 2025. That’s a pace of adoption regulators globally have been racing to keep up with. The UAE is not alone in this move. The US Treasury, the European Central Bank, and the Reserve Bank of India have all issued comparable AI guidance for their own financial sectors recently. The CBUAE’s version notably centers consumer protection and conduct risk as the core framing, rather than treating AI purely as an operational or technical issue.
What This Actually Means for You Practically
If a bank declines your loan application, adjusts your credit limit, or flags your account for fraud review, and AI was meaningfully involved in that call, you now have a clearer, regulator-backed basis to ask for a human explanation. Rather than accepting an opaque automated outcome. This does not guarantee the decision changes. It does mean the bank has a genuine obligation to explain it. And to offer a path to human review specifically, not just customer service boilerplate.
The guidance is not yet legally binding in the strictest enforcement sense. The CBUAE has been explicit that these are flexible, evolving principles, not a fixed rulebook. In practice, treat it the way you would treat any active supervisory framework. Banks that ignore it are exposing themselves to exactly the kind of regulatory friction that shows up in future enforcement action. Customers who cite it directly when requesting a review are on genuinely solid ground.
Sources
* The National: UAE Central Bank issues new guidelines on the use of AI in financial sector — https://www.thenationalnews.com/business/banking/2026/02/23/uae-central-bank-issues-new-guidelines-on-the-use-of-ai-in-financial-sector/
* Khaleej Times: UAE Central Bank issues new rules on AI use in finance to protect consumers — https://www.khaleejtimes.com/business/finance/uae-central-bank-ai-guidelines-financial-sector
* CBUAE Rulebook: guidance note on consumer protection and responsible adoption and use of AI and ML by licensed financial institutions — https://rulebook.centralbank.ae/en/rulebook/guidance-note-consumer-protection-and-responsible-adoption-and-use-artificial-intelligence
Robius.news — Dubai, UAE — 2026 | Built to be first. Built to be trusted.
