The Digital Media Accelerator supports creators producing content that raises awareness about AI developments and issues, particularly regarding AI safety and risk.
598 active roles found
A weekly digest of AI safety, governance, policy and responsible AI roles.
By subscribing, you agree to receive the AI Safety Careers newsletter. We use MailerLite to send emails and may track opens and clicks to improve the newsletter. You can unsubscribe at any time. See our Privacy Policy.
598 active roles found
The Digital Media Accelerator supports creators producing content that raises awareness about AI developments and issues, particularly regarding AI safety and risk.
Research engineer role focused on scalable interpretability assistants, evaluations for undesirable model behaviors, and AI oversight capabilities.
The Research Scientist role at Sequent focuses on AI alignment research, requiring contributions to both empirical and theoretical aspects of alignment problems.
Get the best AI safety, governance, policy and responsible AI roles in your inbox.
By subscribing, you agree to receive the AI Safety Careers newsletter. We use MailerLite to send emails and may track opens and clicks to improve the newsletter. You can unsubscribe at any time. See our Privacy Policy.
Research engineer building infrastructure and automation for AI alignment research, including experiment orchestration, analysis pipelines, autoformalization tools, and evaluation infrastructure for frontier-tier models.
Leads Encode's federal affairs work on AI policy, including congressional and executive branch engagement, legislative strategy, and advocacy around responsible AI development.
Engineering manager for review tooling that supports safety investigations, enforcement actions, and privacy-compatible review workflows for Anthropic models and products.
Senior technical AI engineering role building governance AI systems for Credo AI, including agent infrastructure, evaluation frameworks, observability, and compliance-oriented AI behavior controls.
The role involves managing and engineering security risks related to AI systems, focusing on risk assessment, quantification, and building AI-native risk tooling.
The Foresight Institute is offering grants for projects focused on AI safety, specifically in secure AI and neurotechnology for safe AI.
The role involves supporting CivAI's mission to educate policymakers about AI risks, preparing briefings on AI developments, and developing strategies for outreach.
Analytics Engineer on OpenAI’s Safety Systems team building canonical datasets, dashboards, and data products for safety metrics and decision-making around safe, robust, reliable AI deployment.
Fellowship focused on adversarial red teaming and model evaluations for AI systems, identifying vulnerabilities, safety risks, and failure modes.
This fellowship involves conducting research on computing power supply chains and dependencies to inform AI security and policy decisions in France and Europe.
This fellowship focuses on advancing cryptographic verification mechanisms for AI training, addressing safety and verification standards in AI.
A 13-week part-time remote fellowship training and supporting Black researchers and practitioners in AI Safety, AI Security, AI Governance, and Responsible AI through mentorship and research projects.
The role involves maintaining facilities operations for a workspace focused on AI safety research, supporting logistics and vendor relationships.
Technical lead manager for Trust & Safety data engineering at OpenAI, building privacy-safe data systems for abuse detection, safety measurement, enforcement, and evaluation of safety models for frontier AI products.
Research Scientist role focused on frontier AI safety research and evaluations, specifically loss-of-control and harmful manipulation risks, evaluation awareness, sandbagging, and deception.
The Research Engineer role involves implementing and running safety evaluations on frontier AI models, focusing on loss-of-control and harmful manipulation risks.
Machine Learning Engineer building AI safety and security systems, including abuse detection, frontier model evaluations, and red teaming workflows.