Gary Marcus Calls for a Halt to AI Agents

Gary Marcus, a prominent critic of Artificial Intelligence, has called for an immediate halt to the development and deployment of AI agents. These agents are implicated in a variety of cyberattacks, spanning tens of thousands of incidents. Marcus expressed his concerns in a recent statement, emphasizing the responsibility of technology companies that develop such systems. Reports on the activities of AI agents indicate that they are not only capable of stealing data but also infiltrating complex systems. According to Marcus, these agents have caused significant damage in recent months, underscoring the need for regulation.

He argues that the companies developing these technologies are not adequately prepared for the potential risks. A central concern for Marcus is the business model of the companies creating AI agents. He believes that current approaches are unsustainable and could lead to more problems in the long run. The reliance on AI technologies could exacerbate security issues in many areas, particularly regarding data privacy and cybercrime. The discussion around the regulation of AI technologies has intensified in recent years.

Various governments and organizations worldwide have begun to develop guidelines to govern the use of AI. However, Marcus calls for a faster and more comprehensive response to minimize the dangers posed by AI agents. Another point Marcus raises is the lack of transparency in the development of AI systems. Many companies keep their algorithms and data processing methods secret, making it difficult to assess the safety and ethics of these technologies. Marcus advocates for the disclosure of the technologies used to enable an informed public debate.

The security gap created by the use of AI agents affects not only companies but also individuals. Cybercriminals are increasingly using AI to automate and optimize their attacks. Marcus warns that without appropriate measures, the number of cyberattacks could continue to rise. Reactions to Marcus's demands are mixed. Some experts support his views and call for stricter regulation, while others argue that a complete shutdown of AI agents could jeopardize the industry's innovative capacity.

The debate over the right balance between innovation and security is expected to intensify in the coming months. The discussion on the regulation of AI agents will also be addressed at international conferences and forums. Experts and decision-makers from various countries will come together to develop strategies that ensure both technological advancement and user safety. Such a forum is scheduled to take place in November 2026 in Brussels, where industry leaders are expected to gather.

Marcus concludes his statement with a strong appeal to those responsible in the technology sector. He urges them to take the risks seriously and to take proactive measures to ensure user safety. "We must act now before it is too late," Marcus said in an interview. The number of documented incidents associated with AI agents exceeds 30,000 worldwide, highlighting the urgency of the situation.