OpenAI Safety Fellowship

About the Program

OpenAI is announcing a call for applications to the OpenAI Safety Fellowship, a new pilot program designed to support independent safety and alignment research and develop the next generation of talent in advanced AI systems. The program will run from September 14, 2026, through February 5, 2027.

What You Will Do

• Pursue rigorous, high-impact research on the safety and alignment of advanced AI systems.
• Work closely with OpenAI mentors and engage with a cohort of peers.
• Produce a substantial research output by the end of the program, such as a paper, benchmark, or dataset.

Priority Research Areas

We are looking for applicants interested in safety questions relevant to existing and future systems. Priority areas include:

• Safety evaluation
• Ethics
• Robustness
• Scalable mitigations
• Privacy-preserving safety methods
• Agentic oversight
• High-severity misuse domains

We are especially interested in work that is empirically grounded, technically strong, and relevant to the broader research community.

Requirements

• Applicants from a range of backgrounds are welcome, including computer science, social science, cybersecurity, privacy, HCI, and related fields.
• We prioritize research ability, technical judgment, and execution over specific credentials.
• Reference contacts will be required.

Benefits

The fellowship includes:

• A monthly stipend
• Compute support
• Ongoing mentorship
• API credits and other resources as appropriate (fellows will not have internal system access)

Work Environment

Workspace will be available in Berkeley alongside other fellows at Constellation, though fellows may also work remotely.