Research Scientist • Mid-Level

Who you are

You are an experienced Research Scientist with a strong background in AI safety and risk assessment. You have a deep understanding of the implications of advanced AI models on society and national security, and you are passionate about making AI systems safe and beneficial for users. You thrive in a collaborative environment, working alongside researchers, engineers, and policy experts to tackle complex challenges. Your analytical skills enable you to scope and evaluate potential risks effectively, and you are comfortable running experiments to gather real-world signals.

You have a proven track record of conducting high-impact research, particularly in the context of emerging technologies and their societal implications. You are adept at building research programs and are familiar with the methodologies required to assess risks associated with self-improving and autonomous AI systems. Your ability to communicate complex ideas clearly and effectively makes you a valuable team member.

What you'll do

In this role, you will focus on scoping, evaluating, and defending against societal risks caused by advanced AI models. You will build a comprehensive research program aimed at understanding these emerging risks, which may arise from the increasing integration of powerful AI models in various sectors. You will design evaluations, run experiments, and analyze data to identify potential threats and develop effective defenses.

You will collaborate closely with other members of the Frontier Red Team, contributing to a culture of rigorous research and innovative thinking. Your work will involve engaging with stakeholders across the organization to ensure that the insights gained from your research inform policy and operational decisions. You will also have the opportunity to publish your findings and contribute to the broader discourse on AI safety and ethics.

What we offer

At Anthropic, we provide a supportive and collaborative work environment where you can thrive as a researcher. We offer competitive compensation and benefits, including flexible working hours and generous vacation and parental leave. Our office in San Francisco is designed to foster collaboration and creativity, and we encourage you to apply even if your experience doesn't match every requirement. Join us in our mission to create reliable and interpretable AI systems that benefit society as a whole.