Advancing responsible AI across Europe

openai.com

Every day, millions of people across Europe use OpenAI’s tools to learn, create, work and manage everyday tasks. Our tools also support businesses of all sizes and governments across the region. We believe responsible AI can help drive Europe’s competitiveness and prosperity. As the EU AI Act enters its next phase, we’re sharing how we have strengthened our approach to safety, security, transparency and provenance in line with the EU framework—and how we will continue to evolve our practices as AI advances.

Our longstanding commitment to responsible AI

Our mission is to ensure that artificial general intelligence benefits all of humanity. That mission carries with it an ongoing responsibility to maximize AI’s benefits, broaden access, and manage risks. That is why, for years, we have developed governance approaches and supported workable rules that help people benefit from AI safely, an objective which is central to the EU AI Act. Such rules must be pragmatic, proportionate and risk-based in order to advance governance while supporting innovation. With these principles in mind, we contributed to and endorsed two Codes of Practice: the EU’s General-Purpose AI [GPAI] Code of Practice⁠(opens in a new window) and the Code of Practice on Transparency of AI-Generated Content⁠(opens in a new window), both of which were developed through extensive multi-stakeholder processes and build on work we have advanced across safety, security, transparency, accountability, and provenance.

Building and deploying safe and secure AI

The GPAI Code creates a shared framework for transparency, safety, and security for general-purpose AI models. Our work to support that framework rests on both internal governance and collaboration with external experts, governments, and peer organizations. For years, we have extensively tested our models prior to releasing them, published system cards⁠(opens in a new window) with major releases, brought outside experts into model testing through our Red Teaming Network, and maintained the public Model Spec⁠(opens in a new window) as a window into how we shape model behavior.

We have also continued to strengthen the governance frameworks behind this work. Our Preparedness Framework, in place since 2023 and updated in 2025, sets out how we identify, evaluate, and manage serious risks from advanced AI systems. Our Frontier Governance Framework builds on that foundation and explains how our safety and security practices align with emerging legal requirements, including the EU AI Act’s GPAI Code. Together, these frameworks help translate responsible AI principles into practical decisions about risk assessment, safeguards, model reporting, security, incident response, external expert input, and ongoing updates.

Responsible AI governance also requires work beyond any one company. Through the Frontier Model Forum, collaborations with US CAISI and UK AISI, and work on third-party evaluation practices, we have supported shared safety research, external testing, and clearer evaluation standards across the ecosystem.

Advancing transparency and trust

Our support for the Code of Practice on Transparency of AI-Generated Content builds on years of research, product development, and collaboration to improve provenance for AI-generated media. People should have better context about the content they see online, including whether it was created or edited with AI. OpenAI’s approach to provenance relies on two systems that reinforce each other: Content Credentials (C2PA⁠(opens in a new window)) help content carry detailed context; while SynthID watermarks help preserve a signal when metadata does not survive. We are expanding that work to include audio outputs in addition to images. Consistent with our commitments under the Code, we are working to expand provenance measures for OpenAI’s systems across modalities, including text, as standards and tooling continue to mature. We also aim to support customers and developers building with our models by providing signals, tools, and guidance they can use in meeting their own transparency obligations. Provenance is still an evolving field: metadata can be lost, labels can fail to travel across platforms, and no single signal is perfect. That is why we support a layered approach and continued cooperation across the wider ecosystem.

Adapting governance in practice: the cybersecurity example

Cybersecurity is one area where dynamic and practical governance is especially important. The same capabilities that can help defenders identify and remediate vulnerabilities can also create new misuse risks. Our approach is to reduce misuse while helping legitimate defenders use AI through our Trusted Access for Cyber (TAC) program⁠(opens in a new window), in order to strengthen collective resilience.

We worked to deploy these advanced capabilities and support defenders in Europe. Since launching OpenAI EU Cyber Action Plan in early May 2026, we have worked with EU and national cyber agencies, private sector partners, and critical infrastructure operators, to equip them with the most advanced cyber models, and ultimately strengthen cyber resilience across the continent. This approach aligns with the European Commission’s Action Plan on Cybersecurity and Artificial Intelligence⁠(opens in a new window), which calls for a coordinated approach to address the risks of advanced AI while harnessing its potential to strengthen cyber resilience, including through secure access to advanced AI systems for cybersecurity purposes.

Continuing the work with the entire ecosystem

As implementation of the EU AI Act continues, we will keep strengthening our compliance approach and learning from regulators and the broader ecosystem. As the technology advances, rules must remain flexible enough to adapt and enable people, businesses and organizations to benefit. A pragmatic, proportionate and risk-based approach will be essential if Europe is to realise the benefits of the Intelligence Age.