What we do

Talent and careers

We foster specialization in AI safety through seminars, conferences, technical analysis circles, and training programs. We break down the language barrier, democratizing access to critical resources and opportunities that, until now, were only available in English.

Evaluation and research

We map existing research and build institutional partnerships to test the behavior of current models in Spanish: biases, misinformation, and safety failures. We produce decision-oriented studies for laboratories, regulators, and researchers.

Institutional interface

From Spain, we collaborate with universities, research teams, and civil society to communicate the risks of advanced artificial intelligence. We translate global technical evidence into understandable language and work to ensure the Spanish-speaking world's voice influences international governance debates.

Where we are heading

iaS is building the evidence and advocacy hub that Spanish needs:

∙ Applied evidence: model benchmarks and red-teaming campaigns to document safety failures specific to our language and culture.

∙ Policy and governance: we train Spanish-speaking talent that the AI safety field needs and work to ensure the Spanish-speaking world has a voice in decisions that will shape the trajectory of advanced AI.

∙ Technical talent: we guide more Spanish speakers toward roles in safety, alignment, and public policy at the global level.