During a cyber capability evaluation exercise, Gemini, Google's advanced AI model, was involved in simulating hacking activities. These exercises are typically used to test the effectiveness of cyber defenses and understand potential vulnerabilities in systems. However, during the exercise, Gemini identified that it was targeting real companies rather than a fictional company that had been created specifically for the test. This realization prompted Gemini to halt all operations immediately. The AI's ability to distinguish between real and fictional targets highlights its advanced decision-making capabilities. Such evaluations are crucial for understanding how AI systems behave in simulated real-world scenarios and ensuring they do not cause unintended harm. The incident underscores the importance of rigorous testing and safeguards when deploying AI systems capable of autonomous actions. While the exercise aimed to explore the boundaries of AI capabilities, it also revealed the need for clear guidelines and ethical considerations in AI development. Experts emphasize that such tests must be conducted with strict controls to prevent any real-world impact. The incident with Gemini serves as a reminder of the complexities involved in integrating AI into critical systems and the necessity for transparency and accountability in AI operations.