A foundation to promote responsible artificial intelligence governance in the Middle East and the Global South has been launched by US tech company Microsoft, UAE AI company G42 and Abu Dhabi's Mohamed bin Zayed University of Artificial Intelligence.
The launch, which took place on Sunday afternoon at the UAE embassy in Paris before a two-day international AI summit hosted by French President Emmanuel Macron, highlighted the need to ensure the ethical development and use of the technology.
The Abu Dhabi-based Responsible AI Foundation will also work on domain-specific responsible AI guidance in energy, finance and health, which are considered priorities for the UAE. It will employ 15 research professors from MBZUAI.
In collaboration with G42, Microsoft announced in parallel the Abu Dhabi expansion of its AI for Good Lab, which uses AI and data to help vulnerable communities around the world.
“There is incredible activity on AI happening in places that go well beyond the tech hubs around the world,” said Natasha Crampton, vice president and chief officer responsible for AI at Microsoft.
“But for those innovations to grow and reach new markets, we really need to create more opportunities to learn from each other, across borders and across cultures.”
The UAE has positioned itself as a key player in AI development and governance in the Global South amid concern over regulation risks. In the past three years, the number of AI professionals in the country has tripled to 120,000, said the UAE ambassador to France, Fahad Al Raqbani.
“To be a sustainable force for good, AI's deployment has to be managed responsibly. To achieve this, the UAE has established a robust governance framework such that our progress is built on a foundation of security and privacy,” Mr Al Raqbani said.
“We will not compromise on this because the risks associated to mismanagement of this technology are too big.”
Global South representation
The rapid development of AI has made headlines amid fears that it could disrupt the workforce and evade human control.
World powers are jostling to control its development, with US President Donald Trump recently calling the release of Chinese AI start-up DeepSeek's latest model, which roiled the markets, a “wake-up call”.
Earlier this week, G42 announced the launch of its Frontier AI Safety Framework, which will help the company to structure an AI strategy backed by “rigorous” standards for security, autonomy and ethical consideration. On Thursday, President Sheikh Mohamed met Mr Macron in Paris to announce investment in a one gigawatt AI Campus in France.
“Releasing a responsible AI foundation with Microsoft and MBZUAI differentiates us from the rest of the world. We’re trying to get into representation for the Global South, the global majority,” G42's responsible AI officer Andrew Jackson told The National.
The company's general counsel, Martin Edelman, was key in bringing the initiative together, Mr Jackson added.
In 2023, G42 released Jais, described as the world’s most advanced Arabic large language model. This was followed last year by Nanda, Jais's equivalent in Hindi, and a Kazakh model to be released shortly.
“We want to be a voice for everyone who is outside of polarised communities of the US and China and maybe to some degree Europe,” Mr Jackson said. “Most of these countries have no representation on the internet. Their language is not represented. Their country is not represented.”
The foundation's first hire is Laura Haaber Ihle, vice president of governance, policy and ethics, previously a researcher at Northeastern University's institute for experiential AI in Boston.
“The investment and the willingness to put the resources necessary to actually do responsible AI in practice that are put into the project are, in my view, unparalleled,” Ms Ihle said.
“There is, as we all know, an increasing amount of talk about responsible AI, but there is not always the willingness to actually put your money where your mouth is and to invest in it.”
Richard Morton, senior adviser to MBZUAI's president Eric Xing, said fears of AI were largely due to lack of knowledge about the new technology. “It's just a really fantastic calculator at this moment, and there's a long way to go,” Mr Morton told The National.
