General-Purpose AI (GPAI)
AI models trained on broad data capable of performing diverse tasks across domains.
Definitions (5)
Models intended to perform or be adapted to a wide range of tasks and applications without being narrowly task-specific (including large foundation models); the strategy recognises GPAI as a distinct category requiring particular oversight, standards and potential regulatory attention.
AI models designed and deployed to perform a wide range of tasks across different contexts, adaptable to various use cases rather than being tailored to a single narrow application; treated specially under the AI Act and the Cyprus framework for governance and transparency requirements.
Foundation or general-purpose models covered by Chapter III, Section 2 of the AI Act whose providers must comply with documentation, transparency and disclosure obligations (including model summaries and technical documentation) that apply from 2 August 2025.
Models and systems intended for broad, multi-domain use rather than a single, application-specific task; such models can be adapted or deployed across different contexts and therefore may span multiple risk domains depending on how they are used. The report distinguishes GPAI from application-specific systems to emphasise differing regulatory implications and cross-cutting risks.
AI models that display significant generality and are capable of competently performing a wide range of distinct tasks.