Role Overview

The Researcher, Safety Oversight position at OpenAI is a pivotal role within the Safety Systems team, dedicated to ensuring that advanced AI models are deployed safely and beneficially to society. Located in San Francisco, California, this role involves cutting-edge research in human-AI collaboration, reasoning, robustness, and scalable oversight. OpenAI's mission to build safe AGI requires innovative approaches to maintain oversight over increasingly powerful models.

As a senior researcher, you'll develop AI monitor models, set strategic research directions, and collaborate across teams to uphold the highest safety standards. This position demands passion for AI safety, technical expertise in machine learning, and a commitment to OpenAI's charter. With the rapid evolution of AI capabilities, your work will directly impact how we mitigate risks like misalignment and misuse, ensuring AI benefits humanity universally.

OpenAI invests heavily in novel methods for identifying and addressing safety challenges. You'll thrive here if you're excited about real-world deployment of safe AI systems and eager to advance the field through rigorous research and red-teaming.

Key Responsibilities

In this role, you'll take ownership of critical safety research initiatives. Key responsibilities include:

• Developing and refining AI monitor models to detect known and emerging patterns of misuse and misalignment in frontier AI systems.
• Setting research directions and strategies to enhance the safety, alignment, and robustness of OpenAI's AI models.
• Evaluating and designing red-teaming pipelines to test the end-to-end robustness of safety systems and pinpoint improvement areas.
• Conducting in-depth research to boost models' reasoning capabilities on human values and applying these to real-world safety challenges.
• Coordinating with cross-functional teams such as Trust & Safety, legal, policy, and other research groups to align products with top safety standards.
• Leading projects on scalable oversight techniques tailored for superintelligent AI systems.
• Prototyping system-level interventions for AI misuse detection and mitigation.
• Analyzing learnings from model deployments to refine oversight mechanisms continuously.
• Publishing influential research that advances the broader AI safety community.
• Collaborating on human-AI interaction designs that enable effective safety monitoring at scale.
• Contributing to the strategic vision of safe AI system architecture at OpenAI.
• Mentoring junior researchers and engineers in safety best practices.
• Integrating fairness, bias mitigation, and robustness evaluations into core development pipelines.

These tasks ensure OpenAI remains a leader in responsible AI deployment.

Qualifications

To excel as a Researcher, Safety Oversight at OpenAI, candidates should possess:

• 4+ years in AI safety research, with expertise in RLHF, human-AI collaboration, fairness, and biases.
• Ph.D. or equivalent in computer science, machine learning, or related fields.
• 4+ years of research engineering experience, proficient in Python, PyTorch, or similar.
• Experience with large-scale AI systems and frontier model oversight.
• Strong publication record in AI safety, alignment, or robustness.
• Proven ability to lead research agendas and deliver impactful results.
• Alignment with OpenAI's mission and charter for safe AGI.
• Excellent cross-functional communication skills.
• Hands-on experience in red-teaming and adversarial testing.
• Deep understanding of reasoning, value alignment, and scalable supervision.

San Francisco-based candidates preferred, with relocation support available.

Salary & Benefits

OpenAI offers competitive compensation for this senior role, estimated at $320,000 - $450,000 USD yearly, including base salary, equity, and bonuses. Benefits include comprehensive health coverage, 401(k) matching, unlimited PTO, professional development stipends, catered meals, gym access, relocation assistance, and more. Join a culture prioritizing impact, innovation, and work-life balance.

Why Join OpenAI?

OpenAI is the world's leading AI research organization, pushing boundaries in safe AGI development. Working on Safety Oversight means contributing to humanity's greatest challenge: ensuring powerful AI benefits all. In San Francisco's vibrant tech hub, you'll collaborate with top talent, access state-of-the-art compute resources, and influence global AI policy.

Our commitment to transparency, trust, and diverse perspectives sets us apart. Safety is core to our charter—your research will shape ethical AI deployment. With rapid growth, this role offers unparalleled career advancement in AI safety research.

Experience the thrill of deploying models that transform society while prioritizing safety. OpenAI fosters a supportive environment with employee resource groups, wellness programs, and a mission-driven culture.

How to Apply

Ready to advance AI safety at OpenAI? Submit your resume, cover letter, and links to relevant publications or projects via our careers portal. Highlight your AI safety experience and alignment with our mission. Interviews include technical assessments, research discussions, and team fit evaluations. We prioritize diverse candidates committed to safe AGI.

OpenAI is an equal opportunity employer. Apply now for Researcher, Safety Oversight in San Francisco!