What Is Self-Supervised Learning?

Self-Supervised Learning — a machine-learning approach in which a model learns from unlabelled data by generating its own training signal from the data itself — for example, by predicting masked or withheld parts of the input.

Self-supervised learning is the technique behind most modern foundation models: by predicting hidden portions of huge unlabelled datasets, a model learns general-purpose representations before any task-specific tuning. Its governance significance lies in the scale and provenance of that training data, which raises copyright, privacy, and bias questions that are hard to audit after the fact.

Source: Machine-learning literature