Center for Responsible AI

The Center for Responsible AI (Centro para a IA Responsável) is a Portuguese research consortium dedicated to developing artificial intelligence that is trustworthy, explainable, and aligned with human values. It brings together universities and companies across Portugal — among them NOVA University Lisbon, the Instituto Superior Técnico, the Instituto de Telecomunicações, and the Foundation for Science and Technology, with coordination from researchers including André Martins of Unbabel — supported by Portugal's Recovery and Resilience Plan.

Its flagship project is Amália, launched in July 2026 as Portugal's first government-backed open-source large language model, purpose-built for European Portuguese. Amália is a 9-billion-parameter model built on the European EuroLLM foundation and released fully in the open, with its weights, datasets, and source code all published. The consortium frames the work as a matter of digital sovereignty — giving Portugal a capable, controllable AI system trained on its own language and cultural data rather than depending entirely on models built elsewhere — while emphasizing responsible-AI practices across evaluation, safety, and governance.