Curated roles
Browse roles focused on reducing risks from advanced AI systems, including alignment research, safety engineering, evaluations and related operations.
343 active roles found.
Research engineer intern contributing to AI safety research on security, ethics, alignment, and benchmarking AI risks at a nonprofit focused on high-consequence AI risks.
Conduct research on AI capabilities, risks, and mitigations through benchmarking and alignment assessment.
The role involves establishing risk identification and governance systems for large model security algorithms, developing security capabilities, and researching long-term risks from model improvements.
Lead delivery across an AI Safety portfolio, focusing on model evaluations and strategic implementation in government services.
Faculty role leading an independent research program and teaching/mentoring students, with explicit ability to focus on AI safety research and a college focus that includes AI security governance.
Senior research management role leading Research Managers who support AI safety researchers in an AI alignment and AI governance research program.
Research manager role supporting AI safety researchers and AI safety projects, with explicit involvement in AI alignment and AI governance mentorship and program infrastructure.
Technical lead role owning an AI security hardware workstream focused on hardware architectures for verifying and securing advanced AI systems, with ties to AI safety and compute governance.
Compute infrastructure administrator supporting MATS AI safety fellows and mentors; relevant because the role powers AI alignment and AI governance research infrastructure.
A 16-week remote fellowship conducting peer-reviewed AI safety research in interpretability, governance, evaluation, and alignment.
Research scientist role focused on post-training methods, interpretability, and evaluations to improve frontier AI model safety, robustness, and alignment.
Remote AI red team analyst role focused on adversarial testing of AI models, safety violations, bias, and policy compliance for frontier AI labs.
The Research Lead will develop and lead a research agenda focused on reducing catastrophic risks from advanced AI, mentoring a team and sharing findings through publications and policy briefings.
Software engineer role focused on AI security, including AI red teaming, adversarial machine learning, and building tools and prototypes for AI security applications.
The Research Engineer will develop open-source verification tools and infrastructure for AI-assisted formal verification to ensure bug-free code.
A 5-day London incubator for people building AI safety projects and organizations, with funding for selected teams.
3-6 month fellowship working on an AI safety research project at Constellation's office workspace in the San Francisco Bay Area.
Lead LawZero's Evaluations Team to assess AI capability and safety claims through benchmarks, datasets, adversarial robustness testing, and red-teaming programs.
Research fellowship at CHAI focused on beneficial AI and human-compatible AI research, including alignment-relevant topics like reasoning, decision making, interpretability, reward learning, and adversarial attacks.
This role involves executing operational functions across various domains to support Constellation's mission in AI safety.