The European Commission has issued the final version of the General-Purpose AI Code of Practice.
The Code is designed to help industry comply with the AI Act’s rules on general-purpose AI, which will enter into application on 2 August 2025. The rules become enforceable by the AI Office of the Commission in 2026 for new models and in 2027 for existing models. This aims to ensure that general-purpose AI models placed on the European market, including the most powerful ones, are safe and transparent.
The Code consists of three chapters – the first two apply to all providers of general purpose AI models:
- Transparency;
- Copyright; and
- Safety and Security, which applies to providers of the most advanced models.
As general-purpose AI models underpin many AI systems in the EU, the AI Act helps providers ensure sufficient transparency so that they can integrate these models into their products. The Transparency chapter of the Code contains a Model Documentation Form, which allows providers to document the necessary information in a single place.
The Copyright chapter of the Code offers providers practical solutions to comply with EU copyright law.
Some general-purpose AI models could carry systemic risks, such as risks to fundamental rights and safety, including reducing barriers for the development of chemical or biological weapons, or risks related to loss of control over the model concerned. The AI Act mandates that model providers assess and mitigate these systemic risks. The Safety and Security chapter contains suggested practices for systemic risk management.
Next steps
Once the Code is endorsed by EU member states and the Commission, providers of general-purpose AI models who voluntarily sign the Code will be able to demonstrate compliance with the relevant AI Act obligations by adhering to the Code. The aim is that signatories to the Code will benefit from a reduced administrative burden and increased legal certainty compared with providers proving compliance in other ways.
The Code will be complemented by Commission guidelines on general-purpose AI, which are due to be published before the general purpose AI obligations enter into force. The guidelines will clarify who is in and out of scope of the AI Act’s general-purpose AI rules.
