Agentic Risk Analyst (AI Safety)

TL;DR

Agentic Risk Analyst (AI Safety): Shape a company-wide operating picture for current agentic risk across products and platforms with an accent on system-level risk synthesis across incidents, evaluations, and external developments. Focus on mapping risks to owners and mitigations, running cross-functional intake/review for weak signals, and producing decision-ready assessments with measurable follow-through.

Company

OpenAI is an AI research and deployment company focused on safely deploying powerful AI systems.

What you will do

• Build and maintain a current portfolio of material agentic risks across products, platforms, and emerging capabilities, mapping each risk to workstreams, owners, mitigations, dependencies, decisions, and residual gaps.
• Run a cross-functional intake and review cadence for signals to identify emerging risks and evolving threat patterns, routing findings to the right owners and decision-makers.
• Connect incidents, technical findings, evaluations, and weak signals to broader system-level trends, producing assessments of impact, severity, evidence, uncertainty, priority, and recommended actions.
• Assess how emerging capabilities, product changes, ecosystem developments, and adversary adaptation affect risk priorities and launch readiness, including tracking residual risk after launch.
• Apply and refine frameworks/taxonomies for agentic failure modes, control gaps, abuse patterns, and downstream harms, integrating them into the broader risk program.
• Define and track operating metrics for the agentic-risk program (coverage, ownership, decision latency, mitigation status, closure) and partner with measurement experts on indicators and monitoring.

Requirements

• 7+ years of experience in trust and safety, integrity, security, cyber threat intelligence, AI safety, product risk, strategic intelligence, abuse investigations, or a related field.
• Strong understanding of modern AI systems and agentic architectures, including hands-on experience using, evaluating, or building agentic systems; familiarity with agentic failure modes and misalignment risks.
• Proven ability to analyze how harmful outcomes emerge from interactions between users, products, and technical systems, including experience with cross-functional risk/safety/security/integrity programs and driving follow-through without direct authority.
• Exceptional analytical judgment: identify weak signals, develop/test hypotheses, distinguish signal from noise, and communicate assumptions and confidence clearly under uncertainty.
• Strong technical fluency to translate technical findings into clear risk assessments for engineers, researchers, and security practitioners.
• Comfort synthesizing evidence from multiple sources (investigations, telemetry, evaluations, experiments, dashboards, external research); SQL and Python experience is a plus.

Culture & Benefits

• Work with investigators, engineers, product, policy, safety, and security teams to coordinate decisions and mitigations across the evolving agentic risk landscape.
• Emphasis on turning ambiguous signals into coordinated, measurable follow-through across product, safety, security, policy, and governance.
• Fast-moving environment balancing rigor with pragmatism while building lightweight operating processes and tools.