What Is Fine-Tuning?
Fine-Tuning is the process of further training a pre-trained AI model on a narrower, task-specific dataset to adapt it to a particular use case.
Fine-Tuning — the process of further training a pre-trained AI model on a narrower, task-specific dataset to adapt it to a particular use case.
Fine-tuning takes a general-purpose foundation model and specialises it — for example, adapting a base language model to legal document review or clinical note summarisation. From a governance perspective, fine-tuning is significant because it can introduce new risks (bias from the fine-tuning data, capability changes, degraded safety guardrails) and because it may shift regulatory responsibility: under the EU AI Act, an organisation that substantially modifies a GPAI model can become a provider with its own obligations.
Source: EU AI Act, Article 25; NIST AI 100-1
Plain-language explanation
Fine-tuning takes a general-purpose foundation model and specialises it — for example, adapting a base language model to legal document review or clinical note summarisation. From a governance perspective, fine-tuning is significant because it can introduce new risks (bias from the fine-tuning data, capability changes, degraded safety guardrails) and because it may shift regulatory responsibility: under the EU AI Act, an organisation that substantially modifies a GPAI model can become a provider with its own obligations.
See where you stand on AI governance
Take the free 7-question maturity assessment and get a personalised action plan.
Free assessment — 3 minutes →