AI Safety Researcher (Frontier Model Robustness)

OpenAI

San Francisco, CA • 16m ago • Hybrid

Mid

Hybrid

295,000 - 500,000 USD/yr

  1. Select offer
  2. Add resume
  3. Analyze
  4. Improve
  5. Apply
  6. Auto-send · soon

Match your resume with similar open roles

Upload your resume to automatically match with active openings similar to this role.

PDF up to 10 MB

Your resume is private & secure

Potential salary for AI Safety Researcher (Frontier Model Robustness) role

Improve your resume now

$295,000 – $500,000/year

Quick Facts

  • Role: Researcher focused on frontier AI robustness and containment

Description

You will research and develop methods to secure frontier AI systems. The role covers the full path from threat modeling and experiment design to reproducible results and tested defenses, including code and tool/prototype development. You will partner with Security and infrastructure owners to turn findings into practical, deployed protections.

Responsibilities

  • Design and run experiments to discover failure modes in sandboxes, research infrastructure, and services reachable by models; reproduce vulnerabilities and assess impact.

  • Develop threat models and novel adversarial tests for attacks that cross multiple security boundaries (e.g., sandbox escapes, unauthorized agent communication, access to sensitive data/systems via shared services).

  • Invent, prototype, and evaluate defenses that strengthen isolation and limit blast radius of compromised boundaries; study agent permissions and tool configurations.

  • Build and evaluate model-based safeguards and monitors of model reasoning and actions; test evasion and measure detection/prevention performance.

  • Develop automated adversarial tests and repeatable evaluations to track progress as model capabilities, configurations, and attack strategies change.

  • Partner across Research, Security, and infrastructure teams to prioritize risks and translate research into deployed improvements.

Requirements

  • Hands-on experience conducting research with large language models, including experiment design, execution, and evaluation of model behavior.

  • Ability to convert an open-ended question into a testable hypothesis and produce reproducible, scalable experiments.

  • Ability to design rigorous evaluations, handle unexpected results, and validate findings across different models, environments, and attack strategies.

  • Strong independent coding skills to write and debug code, build research tools, and prototype solutions.

  • Practical security experience relevant to sandboxing, vulnerability research, adversarial testing, and/or detection and monitoring.

  • Ability to collaborate with Security and infrastructure partners to define threat models, interpret findings, and deliver safeguards that can be deployed.

Benefits

  • Opportunity to advance the science of securing frontier AI systems through experimental research and practical defensive systems.

  • Work with cross-functional teams to translate threat findings into layered, infrastructure-aware protections.

See similar jobs

Explore similar jobs and use your resume score to choose where to apply next

See more

Unlock Full Resume Report