AI (Artificial Intelligence) is described here as a very broad research field, but in contemporary usage it often refers to output-generating systems based on machine learning, especially large deep learning models trained on large datasets. In this sense, so-called generative AI optimizes outputs according to given data, objective functions, and evaluation criteria. The process is therefore highly top-down: humans provide the direction by specifying what kind of output they want. At the same time, the internal representations and decision-making processes of such systems are often difficult for humans to interpret, making them effectively black boxes.
In the context of artistic production, AI can be understood as a functional system that returns outputs in response to inputs, but not as a simple mechanism that always produces the same result from the same input. Its uncertainty is treated as a meaningful condition rather than a flaw, because it can help question the conditions of perception, embodiment, and the construction of reality, especially in sound and image-based works. The passages also note a tension in AI-based creation: when outputs are fine-tuned, human arbitrary value judgments inevitably re-enter the process. For that reason, the proper use of AI in creative practice remains something to be examined continuously.