Curated roles
Explore roles in AI red teaming, evaluations, model behavior testing, adversarial assessment and safety measurement.
9 active roles found.
The Red Teaming Fellowship involves conducting adversarial testing of AI systems to identify vulnerabilities and safety risks, executing red teaming exercises, and analyzing model behavior.
The role involves designing and deploying machine learning systems specifically for AI safety and security applications, including model evaluations and adversarial red teaming.
The Research Fellow will conduct research on emerging threats to AI systems, focusing on adversarial red teaming and model evaluations.
The Research Scientist role involves solving research problems related to AI evaluation, robustness, and red teaming of language models.
Lead AI safety research and conduct red teaming for frontier models in sensitive domains.
Develop machine learning-based prototypes and systems for AI security, focusing on red teaming and adversarial machine learning.
Develop machine learning-based prototypes and tools to address AI security challenges, focusing on red teaming and adversarial machine learning.
The role involves leading the Automated Red Teaming effort at OpenAI, focusing on identifying and mitigating AI threats to global security.
The Research Scientist will evaluate and defend against societal risks from advanced AI models, focusing on red teaming and emerging risks.