"What does the government's new AI action plan bring to companies: where are the opportunities, and where are the expectations?" This question is on the agenda of many management teams planning AI investments. The source of the answer is now clear: Presidential Circular No. 2026/9, published in the 18 August 2026 issue of the Resmî Gazete (Türkiye's Official Gazette), brought into force the "Türkiye AI Action Plan (2026-2030)" (Türkiye Yapay Zekâ Eylem Planı). The Circular states that the plan was prepared under the coordination of the Ministry of Industry and Technology (Sanayi ve Teknoloji Bakanlığı), in cooperation with the public sector, the private sector, universities and civil society; it also records that, after the implementation of the 2021-2025 National Artificial Intelligence Strategy, a new action plan was needed.
The Circular assigns duties to public institutions; what it means for the private sector lies in the detail of the plan. In this article, we pull out the topics of direct concern to companies from the 44-page full text published by the Ministry and list the preparation steps that can be taken today.
The framework: four axes, sixteen actions
The plan is built on four axes, "Fark et" (Notice), "İstifade et" (Benefit), "Üret" (Produce) and "Yönet" (Manage), with four actions on each axis, for a total of 16 priority actions. The timeline is divided into two phases: 2026-2027 focuses on establishing governance structures, piloting data spaces and the first public-sector applications, while 2028-2030 brings the scaling of the pilots and investment targets to the fore.
On the governance side, a National AI Council (Ulusal Yapay Zekâ Kurulu), chaired by the President, is envisaged; every year the Council will publish a "National AI Policy Letter" (Ulusal Yapay Zekâ Politika Mektubu). Progress on the actions will be tracked on a publicly accessible Progress and Transparency Portal, and the first progress report is due within 12 months of publication. For companies, this means the calls and the timeline in the plan can be followed from a single place.
Topics that directly concern the private sector
SME vouchers and ready-made solution packages (Action 8). To help SMEs move their AI experiments into lasting products or services, the plan provides for a voucher program. The vouchers will cover software subscriptions, cloud GPU/CPU use, data services and security/compliance testing. SMEs will state their needs on a digital platform and be matched with accredited service providers. The target for the first 12 months is at least 1,000 vouchers. A notable point is the measurement requirement: "before-and-after" measurements will be mandatory in the projects, and the pilots target an average improvement of at least 10% in efficiency or quality over the baseline.
Access to computing power (Action 6). The "GPU for Everyone" (Herkes için GPU) program will offer GPU-hour credits to startups, SMEs and researchers. The aim is for the first call to open within the plan's first 6 months and for at least 2 million GPU-hours a year to be allocated in the first phase; by the end of 2028, this figure is expected to rise to at least 20 million a year.
Data spaces and data sharing (Action 3). Companies and public institutions will be able to share their data in a controlled way, on a voluntary basis, under common rules and standard contracts. For organizations that provide data, incentives such as tax/R&D deductions, GPU/data credits and a "Trusted Data Provider" (Güvenilir Veri Sağlayıcı) status are planned. Companies and consortia selected in health, agriculture, defense and e-commerce will receive support in data, computing and exports under the "National AI Champions" (Ulusal Yapay Zekâ Şampiyonları) program.
Workforce transformation (Action 2). The plan provides for a Workforce Transformation Program (Çalışan Dönüşümü Programı) for the existing workforce, to be designed together with the sectors; businesses that send their employees to the program will be able to benefit from R&D center and technopark (teknokent) incentive mechanisms. The target is for at least 100,000 employees to complete sector-specific competency modules within 24 months.
Public procurement (Action 7). Solutions successfully implemented in the public sector will receive a digital badge, and arrangements will be made for these solutions to be listed in the Tekno Katalog (Techno Catalog) of the Devlet Malzeme Ofisi (State Supply Office). For startups taking part in public procurement for AI projects, the plan takes the techno-venture certificate (teknogirişim belgesi) as the basis. These two items matter for companies thinking of offering solutions to the public sector.
The expectations side: risk-based rules
The plan brings not only support but also rules. Action 4 envisages a national legal and ethical framework that takes the European Union's proportionate risk approach into account. Under it, the following will become standard: simple checklists for low-risk applications, a short-form Algorithmic Impact Assessment (Algoritmik Etki Değerlendirmesi) for medium-risk applications, and, for high-impact systems, an extended assessment, a publicly available summary Model Card and a detailed technical file. Health, education, employment, credit assessment and biometric recognition are among the areas to be treated as high-impact. The aim is for the proportionate risk guide and sector annexes to be published within the plan's first 12 months.
On personal data, the plan says that alignment between KVKK (Türkiye's data-protection law) and the GDPR will be strengthened in the areas of explainability in automated decision-making, personal-data processing in model training and cross-border transfers. Action 16, for its part, envisages setting up regulatory sandboxes, notably in finance, health, energy, mobility and telecommunications, and the publication, within six months, of an action plan on secure cloud use in regulated sectors.
An important distinction: this is an action plan, not a law. The plan states that the legal framework will be established on a general statutory foundation and through secondary legislation to be prepared in consultation with stakeholders. We previously covered the legislative picture in our article on the TBMM (the Turkish Parliament) commission report and the EU-side timeline in our EU AI Act article.
What can companies do today?
The steps below are a general framework derived from the text of the plan; each organization needs to adapt them to its own sector and scale.
- Build a use-case inventory and classify it by risk level. The plan's three-tier structure (low, medium, high-impact) is a good starting point for internal classification. Flag separately any applications that touch areas such as employment, credit or biometrics.
- Start documenting now. For high-impact systems, a Model Card and a technical file are expected. Each model's purpose, data source, performance metrics and known limitations can be put on record today.
- Design your pilots to be measurable. Before-and-after measurement will be mandatory in projects under Action 8. Defining indicators such as time, error rate and cost from the start in your own pilots also makes any future application easier.
- Review your data assets. Participation in data spaces is voluntary; however, knowing which data can be shared is a precondition for benefiting from the incentives. For datasets containing personal data, obtain the opinion of your legal and compliance teams from the outset.
- Match your team plan to the programs. The Workforce Transformation Program and sector-specific certificates are resources that can be included in the internal training plan.
- Follow the timeline. Guides, calls and progress reports will be announced through the Ministry's page and the Transparency Portal. The plan's own timeline puts the targets for the first 6 and 12 months in the foreground.
Conclusion
The Türkiye AI Action Plan sends two messages to the private sector: it aims to ease adoption through tools such as computing power, data and SME vouchers, and, at the same time, to make trustworthy use standard through risk-based rules, impact assessment and documentation. A significant share of the targets is tied to timelines of the first 6, 12 and 24 months; the details will become clear through guides and calls. The soundest preparation therefore runs through steps that will pay off in any scenario, such as inventory, measurement and documentation.
Why logistics experience matters here. The preparation steps above are the working habits of a well-run warehouse or distribution network: count what you hold, keep a record for every unit, define the KPIs before a trial starts, know which data you could share, and plan backward from the dates that matter. The before-and-after measurement that Action 8 makes mandatory for its projects is, in operational terms, a baseline-and-KPI discipline. One of Orient Global's consultants brings more than 30 years in operations, logistics and supply chain: at a technology distribution company of roughly one billion dollars in scale, he ran logistics, warehousing, technical service and administrative operations end to end, building and leading teams of 100+ people and delivering lasting gains in operational performance through KPI-driven programs. He has also managed nationwide warehouse and transport operations. An industrial engineer, he now brings that supply-chain discipline to SMEs through AI and automation. Meet our consultants.
This article is for general information purposes only and is not a substitute for legal advice.
#interimmanagement #forwarddeployedengineering