The researchers conducted a series of experiments in which human pilots were asked to watch videos of AI systems flying a plane. The pilots were then asked to guess what the AI system was doing and why. The results showed that the pilots were often unable to correctly guess the AI system's intentions.
The researchers believe that this is because AI systems make decisions in a very different way than humans. Humans tend to make decisions based on their own experience and knowledge, while AI systems make decisions based on data and algorithms. This difference in decision-making processes can make it difficult for humans to understand what AI systems are doing.
The researchers say that their findings have implications for the future of aviation. As AI systems become more advanced, they will likely play a greater role in flying planes. However, it is important to ensure that humans are able to understand the decisions made by AI systems in order to avoid accidents.
The study's findings suggest that new training methods may be needed for pilots who will be working with AI systems. These training methods should help pilots to understand how AI systems make decisions and how to interpret the information that AI systems provide.
In addition, the researchers say that it is important to develop new ways for AI systems to communicate with humans. This will help to ensure that humans are able to understand the decisions made by AI systems and to make informed decisions about how to interact with them.
