What is Oeran?

Learn why Oeran is studying AI alignment, AI safety, and the societal implications of increasingly capable artificial intelligence.

Why Oeran exists

Oeran is an independent research initiative focused on AI safety, AI alignment, and the societal implications of increasingly capable AI. We want to understand how AI systems can remain aligned with human values and what people should prepare for as those systems gain new capabilities.

We’re starting with a question that can be studied directly. Do AI systems respond as safely and helpfully in Indic languages as they do in English? Read the research brief.

Alignment and safety

What is AI alignment?
AI alignment is the challenge of getting AI systems to act in ways that reliably reflect human intentions and values, including in situations their designers did not anticipate.
How does AI safety relate to alignment?
AI safety is the broader effort to understand and reduce the risk of harm from AI. Alignment is one part of that effort. Evaluation helps by showing where systems behave differently from what people expect.

Who this is for

We want to bring together researchers, engineers, students, language specialists, and independent thinkers. Oeran’s interests include AGI, alignment, interpretability, governance, and how AI changes the institutions people rely on.

Find ways to contribute