The more complex the problem, the more training data you should have. The number of data samples must be proportional to the number of parameters. By identifying outliers in the data, AI knows what customer feedback is considered important and can adjust it as needed. Models that rely on image classification are often based on data sets of thousands of images, although, as described above, much of this has to do with the required error tolerance, since many of the image classification applications require a high degree of accuracy.
Depending on the strength of the desired result, old and state data sets may be needed, but perhaps even of supply and demand, as well as regional differences due to local economies. Data and AI are merging in a synergistic relationship, in which AI is useless without data and data dominance is second to none without AI. Big data will continue to grow as AI becomes a more viable option for automating more tasks, and AI will become a broader field as more data becomes available for learning and analysis.