SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
OpenAI's Global Affairs team is seeking an Applied Risk Standards Specialist to lead the development and advancement of standards for applied AI risks across mental health, youth safety, age assurance, functional efficacy, reliability, and privacy domains.
In this role, you will translate OpenAI's internal research, safety policies, and evaluation methods into technically sound, measurable, and adaptable standards that build public trust and support responsible AI deployment. You will work closely with Safety Systems, Research, Product, Model Policy, Legal, and third-party evaluators to author standards proposals, negotiate requirements, and represent OpenAI in priority standards bodies including ISO/IEC, INCITS, and NIST consortiums.
Key responsibilities include:
- Leading development of applied AI risk standards, with particular focus on evaluation and assessment methodologies
- Translating research and internal policies into credible test methods and assessment criteria grounded in real system behavior
- Bringing emerging external requirements into internal planning and clarifying implications for evaluations, controls, and product assurance
- Collaborating with technical and GRC partners on assessor competence, independence, conflicts of interest, evidence access, and reporting standards
- Contributing to frontier risk-management and independent-assessment standards development
- Leading assigned standards-body workstreams, negotiating proposals, and coordinating internal input to reduce external engagement burden on technical experts
- Maintaining clear approvals, negotiating positions, and implementation handoffs to keep external commitments connected to technical owners
You will own standards drafting, review coordination, negotiation, and assessment-design contributions within your defined mandate. This role does not replace teams responsible for evaluation methods, safety decisions, controls, internal compliance, or implementation.
Qualifications:
- Technical depth in evaluation science, AI evaluations and red teaming, AI safety, or trust and safety, with demonstrated ability to translate expertise into credible standards or assessment methods
- Experience authoring or materially shaping standards, evaluation criteria, control frameworks, or normative technical contributions (not only coordinating participation)
- Ability to translate broad safety objectives into clear requirements and explain what evidence would demonstrate achievement
- Understanding of differences among organizational risk-management processes, model and system evaluations, mitigation testing, and broader assurance claims
- Ability to evaluate sensitive, context-dependent outcomes including uncertainty and variation across users and use cases without overstating what benchmarks or assessments can establish
- Understanding of how assessor competence, independence, incentives, and access affect assessment credibility
- Demonstrated ability to build consensus without sacrificing technical rigor, knowing when to compromise, challenge, or escalate
- Ability to earn trust with technical teams while communicating clearly with lawyers, executives, policymakers, and external standards participants
- Strong writing skills, discretion, and ability to independently move defined workstreams from proposal through review, negotiation, and handoff
