How the plan took effect
Turkey put its 2026-2030 AI Action Plan into effect through a presidential circular dated 18 August 2026. The roadmap announced in June was completed by the August circular, which also assigns duties to public institutions.
The plan contains 16 priority actions across four main pillars, named Recognise, Benefit, Produce and Govern. It gathers 2030 targets into a single document, from education to data centers and from domestic models to private sector investment.
How it differs from the previous document
The chronology needs stating clearly: Turkey previously implemented a separate document covering 2021-2025. The National AI Strategy 2021-2025 aimed to raise AI's share of gross domestic product to 5 percent, bring employment in the field to 50,000 people and place the country in the top 20 of international indices.
The new plan departs from that document in one respect: it foregrounds measurable tasks and concrete implementation headings. Rather than a statement of aims, it sets out a framework for which institution does what.
Recognise: people and data
The first pillar targets AI literacy and human capital directly. The headline numerical targets:
- AI training for 5 million people within two years, through workshops to be set up across all 81 provinces,
- training 10,000 advanced AI specialists and 100,000 application professionals,
- moving at least 2,000 public datasets into the National Data Library for researchers, startups and public institutions.
Those datasets will come from areas such as health, agriculture, defence and e-commerce. The same pillar also provides for handling user rights, privacy and security according to risk levels, with risk-based regulation supported by rules for technology developers and for public institutions.
Benefit: data centers and budget
The second pillar aims to increase the use of AI in public services and production processes. The most concrete item here is infrastructure: Turkey plans to raise total installed data center capacity to at least 1 gigawatt by 2030.
On the budget side, at least a 2 percent share of public investment programmes is to be allocated to AI projects. Small and medium-sized enterprises working in health, energy and smart manufacturing will receive access support through AI vouchers.
How to read it
The most notable aspect of the plan is less the nature of the targets than their form. The previous strategy was largely a statement of direction; this document gives numbers and distributes duties. A measurable target means it becomes possible to ask whether it was met — for a plan, that property is worth more than the size of the ambition.
Some items still need context. A gigawatt of data center capacity is a serious target at Turkey's scale; for comparison, single large AI campuses elsewhere approach that size on their own. The target should therefore be read as building a domestic baseline rather than entering a global race.
The real test will be implementation. The circular distributes duties to institutions, but timetables, budget lines and reporting arrangements sit in the detail of the document; whether those are published regularly will determine whether the plan stays genuinely measurable.