UAE Official Calls for Independent AI Safety Checks Amid Rapid Advances

Countries should abandon expectations that the global artificial intelligence race will decelerate and instead develop independent systems to test advanced models, Faisal Al Bannai, adviser to the UAE President and secretary general of the Advanced Technology Research Council, said at the AI Everything Abu Dhabi summit at Adnec Centre.

Al Bannai rejected proposals to halt technology deployment until comprehensive AI governance formulas are finalized, stating that developers and governments are moving forward rapidly. He argued that organizations developing AI models should not bear the sole responsibility of declaring their systems safe, emphasizing the need for independent oversight particularly when models operate within critical infrastructure.

Rather than conventional static regulations, Al Bannai called for common benchmarks and evolving safety metrics. He noted that scrutiny levels must correspond to risk, contrasting low-impact uses such as drafting emails with high-stakes deployments in healthcare or traffic management.

Addressing sovereign AI capabilities, Al Bannai said nations that deploy external models without visibility into the underlying code or control of model weights risk becoming dependent on closed systems. He stated that the UAE's strategy combines the adoption of top-tier proprietary models with the internal development of open-source models.