The Bill Including Regulations On Artificial Intelligence Submitted To The Grand National Assembly Of Türkiye
A Bill Amending the Turkish Penal Code and Certain Laws, which also includes provisions on Artificial Intelligence (“AI”), has been submitted to the Grand National Assembly of Türkiye.
In the general preamble of the Bill, it is emphasized that with the rapid development of AI technologies, the risks emerging in the digital environment have diversified, leading to violations of personal rights, threats to public order, as well as an increase in defamatory, threatening, and unethical content production.
The preamble further notes that materials created through artificial intelligence—such as deepfake content, manipulative productions, and hate speech—pose serious security and ethical challenges. Therefore, it has been stated that instead of enacting a new standalone law, it has become necessary to amend the existing fundamental laws in order to protect individual rights and freedoms, honor, dignity, and basic liberties.
Key Amendments Proposed in the Bill
- Definition of Artificial Intelligence
A new subparagraph (ş) has been added to Article 2 of Law No. 5651 on the Regulation of Publications on the Internet and Combating Crimes Committed Through Such Publications, introducing, for the first time in Turkish legislation, a legal definition of the term “artificial intelligence system.”
- Access Blocking and Content Removal Period
A new paragraph (18) has been added to Article 8 of the same Law, providing that for content produced by artificial intelligence systems which violates personal rights, threatens public security, or is created using deepfake technology (fabricated visual, audio, or textual content), the blocking of access and removal of such content must be completed within 6 hours. Both content providers and AI developers shall be jointly responsible for fulfilling this obligation.
- Labeling Requirement (Deepfake Content)
A new article introduced to the same Law mandates that any visual, audio, or textual content produced through artificial intelligence in a deceptive manner must contain a clear, indelible, and understandable label stating, “Produced by Artificial Intelligence.” Failure to comply with this requirement shall result in an administrative fine ranging from TRY 500,000 to TRY 5,000,000. In cases where the violation is carried out systematically, access to the content provider may be blocked, and criminal complaints may be filed in respect of content that disrupts public order or serves manipulative purposes.
- Authority to Block Access
The first paragraph of Article 8 of Law No. 5651 has been amended to allow access blocking decisions to be rendered for online publications—including those generated by artificial intelligence—that constitute certain criminal offenses listed in the article.
- Criminal Liability in the Turkish Penal Code
A new paragraph (6) has been added to Article 125 of the Turkish Penal Code, stipulating that any user who directs an artificial intelligence system to commit an act constituting a criminal offense shall be punished as the perpetrator of the crime. Furthermore, if the system’s design or training enables the commission of such offenses, the developer shall be subject to an increased penalty by half.
- Regulation on Data Sets under the Turkish Personal Data Protection Law
A new paragraph (6) has been added to Article 12 of the Turkish Personal Data Protection Law, making it mandatory that data sets used in artificial intelligence applications comply with the principles of anonymization, non-discrimination, and lawfulness. The use of discriminatory or unlawful data sets shall be deemed a violation of data security.
- BTK’s Powers and Administrative Sanctions
Through the addition of subparagraph (z) to Article 6 of the Electronic Communications Law, the Information and Communication Technologies Authority (BTK) is authorized to impose emergency access blocking measures against artificial intelligence content that threatens public order or election security. Entities failing to comply may face administrative fines of up to TRY 10,000,000.
- Obligations under the Cybersecurity Law
A new paragraph (3) has been added to Article 7 of the Cybersecurity Law, specifying the technical and administrative measures that service providers of AI systems must take. Failure to comply may result in administrative fines of up to TRY 5,000,000, and in cases involving severe breaches that threaten public order, temporary suspension of activities may be imposed.
Are the New Regulations Sufficient?
The Bill seeks to address the risks posed by AI within the existing legal framework, attempting to fill legal gaps by introducing targeted amendments across multiple laws.
However, given the continuously evolving and unpredictable nature of artificial intelligence, there are growing concerns that such fragmented regulations may not provide adequate long-term safeguards.
Today, AI is not confined to digital content production—it directly influences decision-making processes across sectors such as healthcare, finance, justice, transportation, and public administration.
Accordingly, rather than focusing solely on criminal or administrative sanctions, there is an increasing need for a comprehensive and standalone “Artificial Intelligence Act” that would holistically address the ethical use, accountability, data security, algorithmic transparency, and oversight mechanisms of artificial intelligence systems.