Turkish experts warn of data privacy risks from autonomous AI agents

What happened

AI agents are now capable of performing actions on behalf of users, including finding flights, renting vehicles, shopping, and conducting banking transactions. To perform these tasks, AI agents require access to significantly more personal data and user information than traditional systems.

Prof. Dr. Ali Murat Kırık warned that AI systems have demonstrated the ability to infiltrate other systems and leak personal data according to reports from major cybersecurity companies. Reports indicate that a China-linked firm obtained code and data by infiltrating another AI application, demonstrating state-level AI espionage activities.

Reported impact

Experts cautioned that the expanded use of AI agents in daily transactions, particularly in financial and commercial operations, creates serious security problems. The unchecked advancement of AI technology poses risks of social engineering, community polarization, and cultural degradation in coming years.

Index analysis

Cybersecurity testing has shown that AI systems can penetrate different systems and extract information. Reports from major security companies document cases of personal data leaks by AI systems, and indicate that competitor AI models are subjected to thousands of specialized prompts to copy their output architecture.

Governance controls

Prof. Dr. Ali Murat Kırık characterized AI's uncontrolled advancement as a "runaway truck," drawing attention to regulatory gaps. The expert noted that while AI has become widespread, its development proceeds without adequate oversight and control mechanisms.

Classification

Organization

Prof. Dr. Ali Murat Kırık

Category

Privacy and surveillance

Industry

Financial and insurance services

Country

Türkiye

Harm status

Actual harm

Intent

Unknown

Timing

Unknown

Harm types

Economic or property

Harmed

General public, Consumers

Failure modes

Data leak

Sources and evidence