OpenAI Says Nearly 80% of UAE Residents Aged 18 to 34 Use ChatGPT Every Week

Solaimani described the UAE and the wider MENA region as unusual in pairing public optimism about AI with heavy investment in infrastructure, energy, computing capacity and talent. He said people in the region did not need to worry about AI taking their jobs, because the region is at the forefront of growth.

The interview came a day after ADNOC, G42 and OpenAI signed an agreement under which ADNOC will become the first energy company to deploy OpenAI Frontier, a platform for coordinating AI agents across business operations.

Usage Figures at ADNOC and in Government

Solaimani said ChatGPT and Codex have been made available to 40,000 people at ADNOC, and he put utilisation at 97%. He also said Abu Dhabi's Department of Government Enablement is piloting an AI upskilling programme for 20,000 government employees.

He named ADNOC, the Department of Government Enablement, Mubadala and Inception as partners OpenAI has worked with in the region for more than three years. His argument was that giving people access to AI tools is not enough unless they are also trained to use them in their work. "Access is not enough. You have to enable and empower your people," he said.

At ADNOC, the work with OpenAI covers field development planning, drilling decisions and reserve estimation, according to Solaimani. The partners are also looking at ways to optimise downstream trading and raise workforce productivity. Another strand is what he called an "intelligence layer", which helps employees query ADNOC's internal data by checking that their questions, labels and terminology match the underlying information. Future work could extend to robotics and offshore operations, he said.

Sovereignty and Local Inference

On AI sovereignty, Solaimani stressed local processing. He said the UAE became the third region in the world, after the United States and Europe, where OpenAI models were available for in-country inference.

He also cited the UAE's 2023 investment in Arabic-language models as a way of drawing in talent and building a lasting technology base. The researchers and engineers who worked on those models are now moving to larger projects, founding startups or joining enterprises, he said.

Solaimani warned against what he called a "false binary" in AI planning, such as choosing between on-premises systems and cloud services, open-source and closed models, or one lab over another. He said the question leaders should be asking is how best to serve their people, and what mix of tools does that.

Why It Matters

The figures are OpenAI's own and have not been independently verified. Still, they show how far the company's Gulf strategy now depends on large institutional deployments. A 40,000-seat rollout at a national oil company, a government training pilot and in-country inference capacity are the kind of commitments that make the UAE one of OpenAI's most important markets outside the United States and Europe. They also matter to regional enterprises weighing their own deployments: OpenAI is presenting training and adoption, not only access to models, as the measure of success.