Intelligent or Not?

Reflection/discussion on robot in context of AI

Our approach can be argued to be a resemblance of machine learning, or having aspects of it, since it involves autonomous decision-making, logical reasoning, and inference to achieve a goal within an uncertain environment. However, it is simple incomparable to any machine learning systems. Unlike actual machine learning algorithms, which adapts and improves its performance over time by identifying patterns in data, our approach relies on predefined rules and logical deductions without any learning component. Still, one could argue that our robot fits machine learning's definition in a limited sense, as the agent updates its internal state (e.g., marking safe or unsafe coordinates) based on new information, resembling a form of "learning." Nevertheless, this "learning" is rigid and does not adapt to different environments or scenarios. Hence, while it can be argued as a machine that "learns," it lacks the dynamic, pattern-recognition abilities that defines most modern machine learning algorithm.

Do you feel any differently regarding the feasibility of AI now that you did this project?

This project did not change our opinions on the feasibility of AI. Just because our robot did not demonstrate intelligence does not mean that having true artificial intelligence is not feasible. This project doesn't scratch the surface of the complexity of AI models that are currently in development, and that artificial general intelligence may still be feasible will algorithms are complex and sophisticated enough.