Final answer:
A key performance indicator (KPI) for a dashboard tracking AI system development would be a table of model evaluation metrics, since they indicate how effectively the AI model is performing and facilitate improvements.
Step-by-step explanation:
The question asks which option is a key performance indicator (KPI) that might be found on a dashboard tracking the development of an AI system. A KPI is a measurable value that demonstrates how effectively a company is achieving key business objectives. For AI systems development, model evaluation metrics are crucial for understanding the performance and progress of the AI models. Therefore, the correct answer to the question is: A table of model evaluation metrics.
Such a table typically includes metrics like accuracy, precision, recall, and F1 score for classification problems, or mean squared error and R-squared for regression problems, among others. These metrics give insights into how well the AI model is making predictions based on the data it has been trained on. This information is utilized to tune, compare, and improve models throughout the development process. Unlike a spreadsheet of the dataset or a collection of graphs that are sprawled over several pages, a table of concise metrics is actionable and directly tied to model performance, making it a quintessential component of a development dashboard.