A study by the Trends Research and Consulting Center has warned of the risks of exploiting artificial intelligence in cyber warfare through the “contamination” of the data it learns from and operates on.
The study, titled “Contamination of Artificial Intelligence: Future Risks of Cyber Warfare,” defined AI contamination as the deliberate introduction of misleading or false information into AI systems with the aim of manipulating their outcomes.
The study indicates that this contamination could have serious consequences for governments, companies, and individuals who increasingly rely on artificial intelligence technology.
The study recommended developing a comprehensive regulatory approach to address the risks of AI contamination, including the “channel” approach proposed by researchers, aiming to guide and secure the development of AI while allowing it to evolve harmoniously with our complex social and technological systems.
It emphasized that current regulations, such as the European AI Act, although positive steps, are still insufficient to counter all the risks of AI contamination, especially against the advanced tactics of adversarial AI.
It concluded by affirming the necessity of preparing to confront the risks of AI contamination as this field continues to evolve and penetrate all aspects of our lives.
