Fine-tuning
Fine-tuning is extra training on a smaller, focused set of examples to adapt a pre-trained model to a task, a style or a field — for example, legal French or a company's support tickets. It changes the model itself, unlike a prompt.
In one line, for a 12-year-old
Fine-tuning is giving an AI extra lessons in one subject so it gets really good at it.
An example
A hospital fine-tunes a model on approved medical summaries so its drafts follow the hospital's format.
Why it matters to people
Fine-tuning on personal or confidential records can leak them later. Organisations should know exactly what data went in and have a privacy assessment before they start.