OpenAI Launches New Safety Fellowship to Advance AI Alignment and Talent
OpenAI is welcoming applicants from diverse backgrounds, including computer science, social science, cybersecurity, and human-computer interaction (HCI).
RMN Digital Research Desk
New Delhi | April 9, 2026
SAN FRANCISCO – OpenAI has announced the launch of the OpenAI Safety Fellowship, a pilot program designed to support independent research into the safety and alignment of advanced artificial intelligence. The initiative aims to foster the next generation of talent by providing external researchers, engineers, and practitioners with the resources to pursue high-impact safety projects.
Focus on Rigorous Research
The fellowship seeks applicants interested in critical safety questions for both current and future AI systems. Key areas of interest for the program include safety evaluation, ethics, robustness, privacy-preserving methods, and agentic oversight. OpenAI is specifically looking for work that is technically strong, empirically grounded, and relevant to the broader research community.
Also Read:
[ OPEN LETTER: Notice to AI Ethics Boards on Bollywood’s Box Office Data Laundering ]
Program Details and Support
Scheduled to run from September 14, 2026, through February 5, 2027, the fellowship offers a comprehensive support package. Participants will receive:
- A monthly stipend and compute support.
- Ongoing mentorship from OpenAI experts.
- API credits and other necessary resources, though fellows will not have access to internal systems.
Fellows will have the option to work remotely or utilize workspace in Berkeley alongside their peers at Constellation. By the end of the program, participants are expected to produce a substantial research output, such as a peer-reviewed paper, a new benchmark, or a dataset.
Application Eligibility
OpenAI is welcoming applicants from diverse backgrounds, including computer science, social science, cybersecurity, and human-computer interaction (HCI). The selection process will prioritize research ability, technical judgment, and execution over specific academic credentials.
Applications are currently open and will close on May 3. OpenAI expects to notify successful candidates by July 25. Interested parties can find the application form and further details regarding compensation on the official OpenAI website.
