Unnoticed, UAE may have curated workable concept for AI regulation

Hindustan Times

17 hours ago

On a recent trip (no secret, to attend the keynote and briefings at Google I/O a few weeks ago), my ears caught the sound of a rather interesting podcast aboard an Emirates A380. Usually these tend to be about food or lifestyle — though the one with Kilian Hennessy is exceptional too — but to have one in which United Arab Emirates' Minister of State for AI, Omar Al Olama talks about his vision of artificial intelligence (AI) regulation, is quite extraordinary. Also because he is the world's first minister for all things AI, appointed in October 2017. His vision is a bit different from everything I've heard from regulators in many countries, including India. Instead of regulation being structured as a restriction, Al Olama's approach lies in a philosophy that hopes to balance innovation with responsibility.
Rather than rushing to implement what may prove to be restrictive regulations, since its still early days considering the contours of the AI space evolve almost every other day, the UAE has positioned itself as what Al Olama calls 'a responsible AI nation.' There are five key tenets to UAE's approach to AI, a working blueprint that many other countries could do well to look closely at. First, the UAE aims to become a global leader in AI by 2031. Secondly, there is belief that AI will be a 'new lifeblood' for both governments and the private sector, transforming various industries. Third, the country envisions a future where individuals have multiple digital assistants, potentially reaching PhD-level capabilities. Fourth, UAE is actively seeking international partnerships and investments to strengthen its position as a global technology hub.
And lastly, there is a willingness to play a key role in the framing of internationally applicable rules to ensure AI security and safety, as well as build frameworks for protection of personal data and privacy.
The fact is simple, Al Olama knows what he is talking about. As far back as in 2019, during a keynote speech at Orchestrating Winning Performance in Dubai, he noted, 'Everyone is talking about AI. Whoever is going to lead in the Artificial Intelligence race will lead the future. This technology will change the world.' Fast forward to 2025, and we have AI which undoubtedly is part of many a workflow for businesses and enterprises, and individuals are learning to trust AI tools more slowly but steadily (whether being careful enough is a debate for another day).
In the past few years, UAE has silently gone about meaning business. But there is a demographic change that needs to happen, sooner rather than later, if the country has to achieve its mission. According to the latest UNESCO data, 0.19% of the UAE's population are working as full time researchers in science, technology, and innovation. In contrast, Korea, which leads this metric, clocks in with 0.69%, followed by Singapore (0.66%) and Norway (0.59%). The US finds 0.42% of its population in these roles, while India clocks in with 0.02% (important to note — this percentage is looking worse than it may be, owing to a much higher comparative baseline population number).
A few days ago, the UAE released a new Arabic-language AI model, called Falcon Arabic, in an effort to make AI locally relevant and accessible. This is something experts too have talked about. India is working on its own LLM, or large language model. As is Singapore. These are just some examples, in the search of an AI that understands local context and sensitivities more than a global model perhaps might. Falcon Arabic has been trained on a dataset spanning Modern Standard Arabic and regional dialects, and Abu Dhabi's Technology Innovation Institute (TII) claims this model matches the performance of models up to 10 times its size. This model may still find it difficult to match up to the likes of Google Gemini, China's DeepSeek and OpenAI as well as Anthropic's fast evolving GPT and Claude models respectively, but it is a start nonetheless.
There is an understanding that isolation, and go-it-alone doesn't work. OpenAI and the UAE are working together for something called the Stargate UAE, a 1-gigawatt AI supercomputing cluster in Abu Dhabi, as part of the AI company's 'OpenAI for Countries' initiative. American tech companies like Oracle, Nvidia, Cisco, and SoftBank, as well as G42, an Emirati artificial intelligence firm, are all part of various AI projects. But it's likely you may have seen excitable headlines about all UAE residents being given free subscriptions to ChatGPT Plus — because that gets the citizens involved.
The US sees this as an opportunity too. Tech companies in the US, presently develop the world's most advanced semiconductor chips. The other side of that coin has the UAE that can offer abundant, low-cost energy that is needed to power enormous AI data centres.
Back to the regulation approach for a moment. The European Union has pursued comprehensive legislation through the AI Act, which categorises AI systems by risk levels and imposes strict requirements on high-risk applications. They have the GDPR as an example that their regulation methods are thorough, as it is the case here too. But then again, it can also be overly complex for tech companies to navigate, and could stifle innovation particularly for companies with limited resources.
The United States has thus far taken a more fragmented approach, with a lack of comprehensive federal legislation, but that could change once the Trump administration gets down to regulating AI. India currently doesn't have rules that regulate AI just yet, but the country is in the process of formulating and implementing policy frameworks to govern various aspects of AI.
The blueprint is there for other countries to look closely at, but the UAE has an advantage of geographical size, its political structure and economic positioning that may be difficult for larger countries to work with — getting a consensus may not be as easy there, as would geographic, economic and human limitations that may come into play. But AI regulation is around the bend, just that we may have differing approaches for the time being.
(Vishal Mathur is the Technology Editor for HT. Tech Tonic is a weekly column that looks at the impact of personal technology on the way we live, and vice-versa. The views expressed are personal.)