Home · Glossary · GPAI (General-Purpose AI)
Enterprise AI glossary · Compliance & Regulation

GPAI (General-Purpose AI)

The EU AI Act category covering general-purpose AI models (foundation LLMs) — Article 53 sets baseline obligations on every provider, Article 55 adds systemic-risk obligations above a 10^25 FLOP training-compute threshold.

Definition

What GPAI (General-Purpose AI) means in practice

GPAI — General-Purpose AI — is the EU AI Act category for foundation models trained on broad data and capable of being adapted to many downstream tasks. Article 53 sets baseline obligations on every GPAI model provider: technical documentation, training-data summaries, copyright compliance, transparency to downstream deployers. Article 55 adds the systemic-risk regime above a 10^25 FLOP training-compute threshold: model evaluations, adversarial testing, incident reporting, cybersecurity for the model artefact itself. Most enterprise customers are not GPAI providers; they're integrators of GPAI models into downstream products. The downstream integration is what Article 25 catches — and the sovereign-deployment advantage is that customer-controlled fine-tunes and customer-owned model artefacts produce a cleaner Article 25 + Article 53 evidence story than vendor-mediated equivalents.

Go deeper
EU AI Act compliance architecture →

All 62 terms, in plain language

Sovereign AI, RAG, agentic AI, IDP, MLOps and the regulations that shape enterprise AI.

Browse the glossary →Talk to an engineer →