Technical Lead, Safety Research

The Safety Research team aims to fundamentally advance our capabilities for precisely implementing robust, safe behavior in AI models and systems. As capabilities continue to advance, it is imperative that our approaches to safety continue to improve and scale to address evolving risks. This is important both for ensuring our systems are robust to prevent harmful misuse as well as ensuring potential misalignment cannot cause harm. We are working on these problems in a way that is grounded in our current models and methods but that generalizes to future systems. We are growing our team to expand our research on methods that will improve safety for AGI and beyond. This will include exploratory research for example, new methods to improve safety common sense and generalizable reasoning, developing new evaluations to elicit or detect misalignment or inner goals of the AI, and new methods to support human oversight of long-running tasks.