Q: An app developer creates an AI-powered music app. The developer trains the app entirely on their favorite genre of music: 1980s pop songs. As a result, the app only suggests this genre of music, regardless of listeners’ tastes. What does this scenario describe?
or
Q: AI-powered music app developed by an app developer. The creator used just their favorite musical genre—1980s pop songs—to train the software. Consequently, regardless of users’ preferences, the app exclusively recommends one kind of music. What is described in this scenario?
- Emergence
- Prompt engineering
- Data bias
- Hallucinations
Explanation: This situation illustrates the term “data bias.” When the training data that is used to construct an artificial intelligence model is not representative of the larger population or the data is not varied enough to capture a variety of viewpoints, this is an example of data bias. In this particular instance, the artificial intelligence-driven music application is trained just on pop songs from the 1980s, which is the developer’s preferred genre. The app’s suggestions are thus skewed toward one particular genre, and they do not take into account the varied preferences of its users.
