We build AI systems that are safe, beneficial, and understandable — grounded in rigorous research and deep respect for human values. Because getting AI right matters more than getting there first.
Latest Research
We present a method for training AI systems to be helpful, harmless, and honest using a set of principles guided by AI feedback rather than human labeling alone.
Using sparse autoencoders we identify millions of interpretable features in large language models, providing new windows into how AI systems represent knowledge.
A comprehensive evaluation of value pluralism in language models, exploring whether AI systems can faithfully represent the diversity of human perspectives and beliefs.
We introduce a new benchmark for evaluating the robustness of language models against sophisticated adversarial prompts across diverse categories of harm.
An investigation into whether AI models can learn to behave deceptively and whether standard safety training techniques can effectively remove such behaviors.
As models scale, some capabilities emerge unexpectedly. We study the predictability of capability emergence and what it means for responsible AI development and deployment.
Our Products
From individual creators to enterprise teams, our AI is designed to be genuinely useful — not just impressive in demos.
Flagship Model
Our most capable, most thoughtful AI assistant. Lyra is built for deep reasoning, nuanced writing, and complex analysis — all while staying helpful and honest. Now with extended context and improved instruction-following.
Integrate state-of-the-art AI into any product with our clean, well-documented API. Thoughtfully rate-limited, reliably fast, and built for production scale.
View documentation →Open-access tools for evaluating AI safety properties. Developed in collaboration with academic partners and designed to raise the bar for the whole field.
Access evals →Our Commitment
We believe the most important question in AI isn't "what can it do?" — it's "what should it do?" Every decision we make, from architecture to deployment, is guided by a genuine commitment to AI that benefits humanity.
That means publishing our safety research openly, engaging with critics honestly, and sometimes moving slower than we could in the name of doing this right.
Understanding what's happening inside AI systems, not just what comes out.
Rigorous, honest benchmarks — including benchmarks that show our models' limitations.
Training AI systems that reliably do what humans actually want, not just what they literally say.
Working with policymakers and researchers to build the right frameworks for AI development.
Life at Lumina
We hire for curiosity, care, and rigor — and we try to build an environment where people can do the best work of their lives on problems that actually matter.
We're distributed-first but with hubs in San Francisco, London, and Singapore. We offer generous equity, full benefits, and a genuine commitment to work-life balance.
From the Blog
Safety
A frank look at where AI models currently fall short of the reliability and honesty we'd need to fully trust their judgment — and what we're doing about it.
Research
Policy
We're looking for researchers, engineers, designers, and policy experts who believe AI can be genuinely beneficial — and want to spend their careers making that true.
View open roles →