Research Scientist

Who you are

You have a strong background in AI safety and model evaluation, with experience in identifying safety-relevant behaviors in machine learning models. Your expertise allows you to design architectures that prioritize safety from the outset, ensuring that models are built with safety in mind. You are familiar with the challenges of monitoring unsafe behaviors and have a proactive approach to mitigating risks during the training process.

You possess a deep understanding of how behaviors emerge in AI systems and can conduct foundational research to measure these behaviors reliably. Your collaborative spirit enables you to work effectively across teams, contributing to a culture of trust and transparency in AI development. You are committed to OpenAI's mission of building safe AGI and are eager to contribute to this important work.

What you'll do

In this role, you will be at the forefront of pioneering safety in AI models before they reach deployment. You will identify safety-relevant behaviors as they first emerge in base models, evaluating and reducing risks without waiting for full-scale training runs. Your work will involve designing architectures and training setups that make safer behavior the default, strengthening models by incorporating richer, earlier safety signals. You will collaborate with various teams within OpenAI's safety ecosystem, ensuring that safety considerations are integrated throughout the model development process.

What we offer

At OpenAI, you will be part of a mission-driven team dedicated to ensuring that AI technologies benefit society. We provide a supportive environment where you can grow your skills and contribute to groundbreaking research in AI safety. You will have the opportunity to work with leading experts in the field and engage in meaningful projects that shape the future of technology. We encourage you to apply even if your experience doesn't match every requirement, as we value diverse perspectives and backgrounds.