Abu Dhabi has launched AiNative, a programme to train 20,000 government employees in artificial intelligence by the end of the year, developed with OpenAI, Microsoft and the Mohamed bin Zayed University of Artificial Intelligence, according to announcements from the emirate’s Department of Government Enablement. The programme teaches staff to use tools including ChatGPT, Codex and Microsoft 365 Copilot in daily government work.
The structure is tiered. Foundation learning journeys are aimed at government employees generally; a public learning path will run through the Tomouh platform; and an executive programme targets senior leaders. A pilot is already operating inside government departments, with broader public access planned for a later phase. The initiative sits inside Abu Dhabi’s AED 13 billion Digital Strategy for 2025–2027 and follows the July rollout of Copilot to some 35,000 employees.
The emirate’s bet is visible in the programme’s premise, stated plainly by its designers: technology and human skills must grow together. Rather than treating AI as a procurement item — licences bought, tools distributed — Abu Dhabi is treating workforce fluency as the deliverable, and measuring success by whether clerks, caseworkers and managers actually change how work gets done.
That is a harder target than it sounds, and governments worldwide are discovering why. Training programmes routinely founder on the gap between a completed course and a changed workflow; the difference is management attention, redesigned processes and permission to abandon old routines. Abu Dhabi’s scale and central coordination give it advantages most governments lack, which makes AiNative a useful test case: if fluency-at-scale works anywhere, it should work here.
The vendors, for their part, gain a reference implementation. Twenty thousand trained government users, on the record, is the kind of deployment OpenAI and Microsoft can cite in every subsequent capital. The experiment begins now; the examination is the year-end target.
There is a competitive geography to the programme as well. The Gulf’s capitals are racing to define themselves as the region’s AI capital, and the race is run less on model announcements — those belong to San Francisco and Beijing — than on the governable assets: sovereign compute, energy, capital and now demonstrably AI-literate state machinery. A trained civil service is a sales asset in that contest, visible to every investor and vendor siting a regional headquarters. Abu Dhabi is, in effect, publishing its bureaucracy’s AI fluency the way other states publish bond ratings. If the training converts to services residents can actually feel, the rating will be earned.
