Türkiye’s Artificial Intelligence Action Plan (2026–2030)

Artificial intelligence has been part of our lives for a long time. However, developments over the past few years have shown that this technology is no longer just a matter for technology companies.

Today, artificial intelligence can be utilized in a bank’s credit assessments, a hospital’s diagnostic processes, a human resources department’s hiring decisions, and even a law firm’s contract reviews. As the scope of this technology’s application expands, a new question arises:

How will these systems be regulated?

The efforts undertaken in many countries around the world over the past few years have been aimed precisely at finding an answer to this question. The Turkiye Artificial Intelligence Action Plan (2026–2030), announced by President Recep Tayyip Erdoğan at the Turkiye Artificial Intelligence Summit in Istanbul on June 13, 2026, is one of the most concrete steps in this quest within Turkiye.

At first glance, the document may appear to be nothing more than a policy paper. There is no new law; nor is there a binding regulation that imposes direct obligations on companies.

However, profound transformations in technology law often begin not with laws, but with strategic documents. The regulation we now know as the European Union’s Artificial Intelligence Act (AI Act) also took shape following years of policy work, expert reports, and strategic documents.

For this reason, it would be insufficient to view the Action Plan merely as a vision document. For a legal professional, the key question is: What message is Turkiye sending about the future through this document?

To understand the answer to this question, we must first examine the path Turkiye has followed in the field of artificial intelligence over the past few years.

Turkiye’s AI Journey: A Chronological Overview

2021 – National Artificial Intelligence Strategy

Turkiye’s first comprehensive policy document in this field was the National Artificial Intelligence Strategy (2021–2025), published in 2021. Prepared by the Presidential Office for Digital Transformation and the Ministry of Industry and Technology, the document outlined objectives such as training AI experts, increasing research capacity, expanding the use of AI in public services, developing data and technical infrastructure, and strengthening international collaborations.

During this period, Turkiye’s approach focused primarily on the development and widespread adoption of artificial intelligence; regulation and oversight were not yet at the center of the agenda.

Institutionalization of the Ecosystem (Post-2021)

After 2021, the number of institutions and platforms working in the field of artificial intelligence increased significantly. The prominent entities include:

• Presidential Office for Digital Transformation — served as the central body responsible for coordinating digital transformation policies. (cbddo.gov.tr)

• TÜBİTAK Artificial Intelligence Institute — has become one of the most important publicly funded research centers, focusing primarily on machine learning, big data, natural language processing, decision support systems, and public projects. (yte.org.tr)

• TÜBİTAK BİLGEM — has recently stood out particularly for its Turkish large language model and generative AI projects. (bilgem.tubitak.gov.tr)

• Turkiye Artificial Intelligence Initiative (TRAI) — one of the leading platforms bringing together startups, investors, and corporate companies. (turkiye.ai)

• Artificial Intelligence Vision Platform — established in 2026 by the Ministry of Industry and Technology to oversee the preparation of the Action Plan. (yapayzekavizyonu.sanayi.gov.tr)

A Game-Changing Regulation: The EU AI Act

To put developments in Turkiye into context, it is necessary to look at the steps taken by the European Union. This is because the most significant development shaping the direction of AI law today is, without a doubt, the AI Act.

Adopted in 2024 and phased in starting in August 2024, the AI Act is recognized as the world’s first comprehensive artificial intelligence regulation. Its significance does not stem solely from the fact that it will be applied in Europe: considering the “Brussels Effect” that the GDPR created in data protection law, the AI Act is also expected to become a global standard.

The underlying logic of the AI Act is straightforward: not all artificial intelligence systems carry the same level of risk. A customer service chatbot cannot be treated the same as a system that evaluates a person’s loan application or hiring decision. For this reason, systems are classified as unacceptable risk, high risk, limited risk, and low risk; as the risk increases, so do the obligations of developers and users. Systems used in hiring, credit assessment, education, and critical infrastructure are generally classified as high-risk and are subject to obligations such as transparency, human oversight, and compliance assessments.

This risk-based approach has already begun to influence regulatory efforts in many countries, from Canada to Brazil and from the United Kingdom to Japan. Turkiye is no exception to this trend.

The Turkish Grand National Assembly Puts Artificial Intelligence on Its Agenda (2024)

2024 marked a turning point for Turkiye.

Artificial Intelligence Research Commission. Pursuant to a decision published in the Official Gazette on October 5, 2024, an Artificial Intelligence Research Commission was established within the TBMM. The commission was tasked with examining the economic impacts of artificial intelligence, its legal risks, its implications for public policy, and international practices. This marked the first time artificial intelligence had become the subject of institutional work at the legislative level. The commission completed its work and shared its report with the public in the first half of 2026.

The First Artificial Intelligence Bill. During the same period, Turkiye’s first Artificial Intelligence Bill was also submitted to the Turkish Grand National Assembly (June 25, 2024). Although the bill did not become law, it was significant for the legal community: for the first time in Turkish law, concepts such as high-risk AI systems, transparency, accountability, human oversight, conformity assessment, and the liability of AI providers began to be systematically discussed. An examination of the bill’s language and approach reveals that it was clearly influenced by the AI Act.

In the period following this initial bill, the issue remained on the agenda; throughout 2025, new legislative proposals related to artificial intelligence continued to be submitted to the Turkish Grand National Assembly (TBMM). This demonstrates that legislative interest is not a one-time initiative but a trend that has gained momentum.

What Does the 2026 Action Plan Entail?

Following all these developments, the Turkiye Artificial Intelligence Action Plan was announced in 2026. Therefore, it would be incorrect to evaluate the document in isolation; the Plan is, in fact, a natural continuation of the 2021 National Artificial Intelligence Strategy, the work of the TBMM, the legislative proposals, and global developments following the AI Act.

The Plan is built around four core pillars — each featuring four complementary actions:

• Awareness — AI literacy, educational programs, and human resources. (The Plan aims to train 5 million citizens over two years through workshops to be established in all 81 provinces.)

• Leverage — the use of artificial intelligence in public services, increased efficiency, and digital transformation.

• Develop — Turkish large language models, domestic solutions, research funding, and computing infrastructure.

• Govern — risk-based governance, regulatory sandboxes, data governance, and compliance with international standards.

From a legal perspective, the last point is particularly noteworthy. Indeed, the Plan’s introduction explicitly states that a regulatory framework will be established that protects user rights, privacy, and security while fostering innovation, provides predictability for investors, and is based on a “proportionate risk approach.”

Is Turkiye Moving Toward an AI Act?

As of today, there is no law in Turkiye similar to the AI Act currently in force. However, the terminology used is noteworthy. The concepts of proportionate/risk-based governance, trustworthy AI, regulatory sandboxes, transparency, and data governance included in the Action Plan largely align with the fundamental approach of the AI Act.

The table below summarizes the aspects of the plan that align with the AI Act but have not yet been addressed:

Turkiye’s current approach appears to be similar to the EU AI Act. This suggests that Turkiye is more likely to move toward a local regulatory framework informed by developments in the European Union rather than adopting a model built from scratch that is entirely different.

Where Does Convergence End, and Where Does Divergence Begin?

However, “aligning with the AI Act” does not mean “copying the AI Act.” Conceptual similarities can lead to significant divergences in institutional architecture.

The European Union has established an integrated framework consisting of a single, comprehensive regulation; clear risk categories; a conformity assessment process; specific rules for general-purpose AI (GPAI) models; and market oversight authorities. Turkiye’s legislative efforts to date, however, reflect a different approach: rather than enacting a standalone “Artificial Intelligence Law,” there is a tendency to amend existing laws — particularly the Turkish Penal Code — to delegate oversight to existing institutions such as the BTK, and to focus on concrete risks such as deepfakes and content liability.

Therefore, a likely scenario is a hybrid model that borrows the risk language and conceptual framework of the AI Act but charts its own course regarding institutional structure, enforcement models, and issues such as GPAI. In short, while Turkiye may adopt Europe’s “what,” it may take a unique approach to the “how.”

What Should We Expect Next?

Looking at the current picture, the trajectory is quite clear:

This timeline demonstrates that Turkiye is striving not only to produce technology but also to establish a governance and regulatory model. In the coming years, it would come as no surprise to see AI governance standards, rules regarding high-risk systems, transparency obligations, oversight mechanisms, sector-specific regulations, and ultimately a potential Artificial Intelligence Law come to the forefront.

A realistic forecast regarding the timeline is as follows: in the short to medium term (approximately 12–24 months), issues such as transparency and deepfake labeling, along with sector-specific regulations and targeted amendments to existing laws, may take center stage; a comprehensive and standalone AI Act, however, may take longer to mature and the timeline remains uncertain.

This transformation directly affects not only the government but also public institutions and private companies.

Expectations for Public Institutions and Private Companies

Although the Action Plan does not impose direct obligations, it presents both new opportunities and implicit expectations for public institutions and private companies. It is helpful to consider these two aspects separately.

Open Data and Infrastructure: New Opportunities

One of the key topics highlighted in the plan is open data. Under the plan, the goal is to make at least 2,000 public datasets — primarily in the fields of health, agriculture, defense, and e-commerce — available to researchers, entrepreneurs, and local developers through the National Data Library to be established.

So what does “open data” mean? In short: it means making data — which until now has been scattered across public institutions and often inaccessible — reusable within a specific framework of rules. Since the success of an AI model depends largely on the quality and breadth of the data used to train it, access to high-quality, extensive, and contextually relevant public datasets could become a critical competitive advantage, particularly for companies developing Turkish-language solutions.

In addition to open data, the plan envisions AI vouchers for SMEs, “AI growth zones” with ready energy and infrastructure, the public sector becoming the first adopter of domestic solutions, and a large-scale investment drive — led by the private sector — in data center, cloud, and AI infrastructure. These measures translate into concrete business and growth opportunities for the private sector.

However, “Open” Data Does Not Mean “Free” Data

From a legal perspective, the term “open” does not mean “free for any use.” For companies using publicly available datasets, the following issues stand out:

• Personal data and the Personal Data Protection Law (KVKK): if the data contains personal information, being “open” does not eliminate KVKK obligations; the principles of anonymization, legal basis, and purpose limitation remain in effect.

• Risk of re-identification: the possibility of re-identifying individuals by combining datasets considered anonymous with other sources poses a legal and reputational risk in and of itself.

• Terms of use and licenses: open data is often subject to specific licenses and terms of use; rules regarding commercial use, attribution, and redistribution must be observed.

• Sensitive sectors: sensitive data — particularly health data — is subject to additional protections and restrictions, even if it is “open.”

What Is Expected in Return

While the government provides the data, infrastructure, and incentives, it also expects institutions and companies to manage artificial intelligence responsibly. As we move toward a risk-based framework, maturity in data governance, transparency, human oversight, and accountability will become increasingly critical.

Today, many companies use artificial intelligence at various stages of their operations; however, most organizations have not yet been able to provide systematic answers to the following questions:

• Which AI systems are we using?

• What data do these systems use?

• To what extent are the decisions they produce explainable?

• What is the level of human oversight?

• How are erroneous outputs managed?

One of the most significant impacts of the Action Plan may be that it forces companies to confront these questions earlier. This is because the global trend is no longer just about developing AI, but about being able to manage it — and it appears that Turkiye does not want to be left out of this transformation.

Recommended Actions

The phased roadmap we recommend for organizations that want to prepare for the above questions starting today aims to develop a proactive compliance capacity even during a period of regulatory uncertainty.

Short Term (0–12 Months)

• Developing an AI Governance Policy

• Creating an AI inventory

• Mapping AI usage

• Conducting a review of suppliers and third-party AI

Medium Term (12–24 Months)

• Conducting AI risk assessments

• Establishing model validation processes

• Establishing AI oversight mechanisms

• Conducting KVKK compliance checks

Long Term (24–48 Months)

• Preparing for regulations similar to the AI Act

• Participation in regulatory sandbox programs

• Integration of domestic large language models (LLMs)

• Establishment of regulatory reporting systems

Conclusion

It would be an oversimplification to view Turkiye’s Artificial Intelligence Action Plan merely as a new policy document. The document has emerged as a synthesis of the strategies, institutional structures, parliamentary efforts, and international developments that have taken shape over the past five years.

Turkiye does not yet have a comprehensive Artificial Intelligence Law. However, it is clear at this stage that the country aims not only to develop artificial intelligence but also to regulate and manage it.

For this reason, the 2026 Action Plan should be viewed not as a final destination, but as a significant milestone on the path toward Turkiye’s potential artificial intelligence regulatory framework.