UAE Charter for the Development and Use of Artificial Intelligence

What it is

The UAE Charter for the Development and Use of Artificial Intelligence is the United Arab Emirates’ national statement of how AI should be developed and used. The UAE Cabinet approved it as an ethical charter of 12 principles, aimed at preserving respect for human values and ensuring fair and safe treatment for all members of society. The UAE legislation portal lists its issue date as 10 June 2024.

The Charter sits under the UAE Strategy for Artificial Intelligence. Its stated objectives are to use AI ethically and responsibly, protect privacy and data security, balance technological progress with social values, encourage innovation and economic growth, raise awareness and education about AI, improve transparency and accountability, and strengthen the UAE’s standing in technology and innovation.

It builds on earlier UAE work, including the UAE AI Ethics Principles and Guidelines published in December 2022.

Who it applies to

The Charter addresses the development and use of AI in the UAE generally, rather than one sector or one kind of organisation. It is a policy, not legislation, so no organisation is legally bound by it and there is no enforcement mechanism.

It is most useful in three situations. A government body can use it as the reference point for its own AI use. A supplier selling AI-enabled products or services into the UAE can use it to frame answers about how its AI is governed. And a multinational operating in the UAE can map its global AI governance programme to it to show local stakeholders that national expectations are covered.

What it requires

The Charter asks organisations to develop and use AI in line with 12 principles. The wording below is Aicura’s summary of each.

1. Strengthening human-machine ties. AI should support a beneficial relationship with people, with human well-being and progress as the priority.
2. Safety. AI systems should meet high safety standards, and systems that pose risks should be modified or withdrawn.
3. Algorithmic bias. Bias in algorithms should be addressed so that AI is fair, inclusive, accessible and free from discrimination.
4. Data privacy. Data can support AI innovation, but the privacy of community members remains a top priority.
5. Transparency. People should be able to understand how AI systems operate and reach decisions.
6. Human oversight. Human judgement and oversight of AI remain essential, so that errors and biases can be caught and corrected.
7. Governance and accountability. Organisations should take a proactive approach to governing AI so that it is used ethically and transparently.
8. Technological excellence. The UAE aims for global leadership in AI to drive innovation, competitiveness and sustainable progress.
9. Human commitment. AI should serve the public good, protect fundamental rights and keep human values at the centre.
10. Peaceful coexistence with AI. AI should improve community well-being without compromising human security or fundamental rights.
11. Promoting AI awareness for an inclusive future. Everyone should be able to benefit from AI, with equitable access across society.
12. Commitment to treaties and applicable laws. AI should be developed and used in line with international treaties and local laws.

Most of these translate into familiar governance practice. Safety, bias, privacy and transparency become risk and impact assessments with documented mitigations. Human oversight becomes a recorded decision about how much autonomy each system has and who can intervene. Governance and accountability become named owners, a policy set and a record that shows it being followed. Principles 8 and 11 are national goals rather than organisational duties.

Key dates

• December 2022. UAE AI Ethics Principles and Guidelines published.
• 10 June 2024. Charter issue date on the UAE legislation portal.
• June 2026. The UAE announces a federal Artificial Intelligence and Data Authority, reporting to the Cabinet. It consolidates the AI Office, the UAE Data Office and the digital government sector of the Telecommunications and Digital Government Regulatory Authority, and is to set data and AI standards for federal entities.

How it relates to other frameworks

The Charter’s principles line up closely with the OECD AI Principles and with other national principle sets, such as Singapore’s Model AI Governance Framework and Australia’s AI Ethics Principles. Like them, it says what good looks like without saying how to get there. ISO/IEC 42001 and the NIST AI RMF supply the how, with management system requirements and risk management functions that turn principles into controls and records.

Within the UAE, the Charter is the national baseline, while binding obligations come from elsewhere. Inside the Dubai International Financial Centre, DIFC Regulation 10 sets enforceable rules for AI systems that process personal data. Financial institutions also face expectations from their own regulators. An organisation that can show the Charter’s principles in its policies and assessments will usually find it has much of the groundwork those binding rules ask for.

A note on this page

This is Aicura’s reading of the Charter and the official UAE descriptions of it. It is not legal advice and it is not the Charter itself. Read the primary source on the UAE government portal, and take advice on how UAE expectations apply to your organisation.