Türkiye minister: AI authority can be delegated, not responsibility
09:27, 08/11/2026, Sunday
AA
File photo
Industry Minister Mehmet Fatih Kacir told the TRT World Forum that artificial intelligence systems require strict human accountability, warning that the concentration of computing power among a handful of corporations poses structural risks to safety and transparency while calling for a human-centered regulatory approach.
Human-centered AI governance
Türkiye's Industry and Technology Minister Mehmet Fatih Kacir on Saturday warned that artificial intelligence systems must operate within strict accountability frameworks, telling the TRT World Forum that technological advancement cannot come at the expense of human responsibility. Speaking at the panel "Navigating Responsibility in the Age of AI," Kacir said increasingly autonomous systems raise fundamental questions about transparency and safety. "We can delegate authority to AI, but we cannot delegate responsibility," he said, highlighting risks including algorithmic discrimination, cybersecurity vulnerabilities and deepfakes.
Market concentration risks
Kacir noted that the global AI landscape remains dominated by a small cohort of private actors, with more than 90 percent of frontier models developed by industry and just five companies controlling an estimated 70 percent of worldwide computing power. This concentration, he warned, creates structural vulnerabilities that transcend national borders and complicate regulatory oversight.
2030 national roadmap
Outlining Ankara's strategic response, Kacir said Türkiye's 2026-2030 Artificial Intelligence Action Plan aims to establish the country among leading technology nations. The initiative targets AI literacy training for 5 million citizens, the cultivation of 10,000 advanced specialists and 100,000 professionals, and the release of at least 2,000 public datasets. Türkiye also seeks to expand domestic data center capacity to one gigawatt by 2030 while attracting $10 billion in private investment across cloud infrastructure and computing facilities.
He announced that Türkiye's domestic large language model, BILGE, will soon enter open testing for developers, bolstering technological sovereignty. Regulatory sandboxes and a dedicated AI Ethics Board are also planned to address privacy, discrimination and child safety concerns, Kacir added.
Multilateral cooperation
Türkiye will advance its regulatory framework through the G20, OECD, UNESCO and the Organization of Turkic States, Kacir said. The minister noted that Ankara also intends to contribute to United Nations initiatives aimed at establishing inclusive governance structures ahead of the 2026 implementation window for the national action plan.