Chain-of-thought Interpretability Team - Research Science and Engineering

Monitoring models for misaligned or dangerous behavior is a crucial mitigation to our mission of bringing safe artificial general intelligence to the world. We have long been excited by the prospect of monitoring our models’ latent thinking in addition to outputs that users see; however, until recently that latent thinking has only been available in the form of activations – large blocks of illegible numbers from which we have only been able to extract simple concepts. With the advent of models that rely heavily on chain-of-thought (CoT) reasoning to solve complex tasks, we now have access to some of the models’ internal thinking in a far more legible form and could allow us to monitor their latent thinking for more complex behavior.

The Chain-of-thought Interpretability Team, part of OpenAI’s Science of Deep Learning team, is working on technical approaches to determine whether model CoTs are monitorable, i.e. when they are faithful and legible, and what interventions may improve or degrade monitorability.

About the Role

In this role, you will develop innovative machine learning techniques and collaborate with peers across the organization to advance this critical pillar of OpenAI’s mission. We are looking for people who want to discover simple, generalizable ideas that work well even at large scale, and form part of a broader research vision that unifies the entire company. We are further looking for individuals with solid engineering skills, that can write bug-free ML code, and can work in the complex code bases behind our state-of-the art AI systems.

You will thrive in this role if you

• Are excited about OpenAI's mission and eager to move the needle on a critical component of building safe, beneficial AGI

• Are eager to study AI safety through a scientific lens

• Have a background in statistical machine learning, physics, mathematics, or another theoretically and empirically rigorous field

• Are passionate about building, running, and studying AI systems at the largest scales and at the forefront of the field

• Enjoy fast-paced and collaborative research environments focused on achieving the impossible

About OpenAI

OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity. We push the boundaries of the capabilities of AI systems and seek to safely deploy them to the world through our products. AI is an extremely powerful tool that must be created with safety and human needs at its core, and to achieve our mission, we must encompass and value the many different perspectives, voices, and experiences that form the full spectrum of humanity. 

We are an equal opportunity employer and do not discriminate on the basis of race, religion, national origin, gender, sexual orientation, age, veteran status, disability or any other legally protected status. 

OpenAI Affirmative Action and Equal Employment Opportunity Policy Statement

For US Based Candidates: Pursuant to the San Francisco Fair Chance Ordinance, we will consider qualified applicants with arrest and conviction records.

We are committed to providing reasonable accommodations to applicants with disabilities, and requests can be made via this link.

OpenAI Global Applicant Privacy Policy

At OpenAI, we believe artificial intelligence has the potential to help people solve immense global challenges, and we want the upside of AI to be widely shared. Join us in shaping the future of technology.