Frontier AI Research · San Francisco
The Intelligence and Investigations team seeks to rapidly identify and mitigate abuse and strategic risks to ensure a safe online ecosystem. We are dedicated to identifying emerging abuse trends, analyzing risks, and working with our internal and external partners to implement effective mitigation strategies to protect against misuse. Our efforts contribute to OpenAI's overarching goal of developing AI that benefits humanity.
The Strategic Intelligence & Analysis (SIA) team provides safety intelligence for OpenAI’s products by monitoring, analyzing, and forecasting real-world abuse, geopolitical risks, and strategic threats. Our work informs safety mitigations, product decisions, and partnerships, ensuring OpenAI’s tools are deployed securely and responsibly across critical sectors.
As an Agentic Risk Analyst, you will shape OpenAI’s operating picture for current agentic risk across products and platforms. You will bring a strategic, system-level perspective to current risks, connecting individual incidents, technical findings, abuse patterns, and external developments to relevant workstreams, mitigations, owners, dependencies, and residual gaps. You will analyze how risks emerge through autonomy, multi-step task execution, tool use, memory, retrieval, connectors, computer-use capabilities, and multi-agent workflows, with a particular focus on both adversarial misuse and unintended system behavior. By synthesizing signals from investigations, evaluations, red teaming, security reviews, product launches, external research, and real-world incidents, you will maintain a current view of material risks and evolving threat patterns. Your work will help turn complex and often ambiguous signals into coordinated decisions and measurable follow-through across product, safety, security, policy, and governance teams.
You will work closely with investigators, engineers, product, policy, safety, and security teams, and measurement and forecasting experts who lead longer-horizon risk discovery and scenario work to maintain a shared operating picture of current risks, mitigation priorities, owners, and dependencies. This is an opportunity to help shape how OpenAI coordinates decisions and follow-through across the evolving risk landscape of increasingly capable agentic systems, ensuring that safety decisions keep pace with rapidly advancing technology.
Build and maintain a current, company-wide portfolio of material agentic risks across OpenAI’s products, platforms, and emerging capabilities, mapping each risk to relevant workstreams, owners, mitigations, dependencies, decisions, and residual gaps.
Run a cross-functional intake and review cadence for signals from across OpenAI and the broader ecosystem to identify emerging risks, evolving threat patterns, and important shifts in the agentic risk landscape, routing findings to the right owners and decision-makers
Connect individual incidents, technical findings, evaluations, and weak signals to broader system-level trends, producing clear assessments of impact, severity, evidence, uncertainty, priority, and recommended action.
Assess how emerging capabilities, product changes, ecosystem developments, and adversary adaptation may affect current risk priorities and launch readiness, surfacing risks that are unowned, stalled, or under-mitigated for decision and escalation, and tracking residual risk after launch.
In partnership with colleagues who lead horizon scanning, use relevant external developments across AI safety and security research, public incidents, adversarial activity, industry standards, emerging technologies, and competitor products, as inputs to current risk prioritization and mitigation decisions at OpenAI.
Apply and refine practical frameworks and taxonomies for current agentic failure modes, control gaps, abuse patterns, and potential downstream harms across products, deployment environments, and user workflows, integrating them into the broader risk program.
Produce concise, decision-ready assessments that communicate key findings, assumptions, confidence levels, competing hypotheses, and prioritized recommendations for product, safety, security, policy, and leadership stakeholders, with accountable owners and next review points.
Define and track operating metrics for the agentic-risk program, including coverage, ownership, decision latency, mitigation status, and closure and partner with measurement experts on risk indicators and monitoring capabilities.
Partner with colleagues who lead evaluations, scenario analyses, tabletop exercises, red-team campaigns, and investigations focused on agentic systems, using their findings to drive actionable mitigations, accountable ownership, and measurable follow-through.
Have significant experience—typically 7+ years—in trust and safety, integrity, security, cyber threat intelligence, AI safety, product risk, strategic intelligence, abuse investigations, or a related field.
Have a strong understanding of modern AI systems and agentic architectures, including hands-on experience using, evaluating, or building agentic systems. You are familiar with AI safety concepts, agentic failure modes, and misalignment risks, and can reason about how agent objectives, incentives, environments, and system design choices may produce unintended or harmful outcomes.
Have demonstrated experience analyzing how harmful outcomes emerge from interactions between users, products, and technical systems, and can assess how increasingly capable agentic systems may amplify, automate, or transform existing abuse vectors such as fraud, scams, social engineering, coordinated influence operations, cyber abuse, or other forms of platform misuse. You have also helped operate a cross-functional risk, safety, security, or integrity program, clarifying ownership, resolving dependencies, escalating gaps, and driving follow-through without direct authority.
Have exceptional analytical judgment and experience producing assessments under uncertainty. You can identify weak signals, develop and test hypotheses, distinguish signal from noise, communicate assumptions and confidence levels clearly, and update your conclusions as new evidence becomes available.
Possess strong technical fluency and are comfortable engaging deeply with engineers, researchers, and security practitioners. You do not need to be an engineer, but you should be able to translate technical findings into clear risk assessments.
Are comfortable synthesizing data and evidence from multiple sources to manage a current risk portfolio, including investigations, telemetry, evaluations, experiments, dashboards, and external research. Experience using SQL, Python, and analytical tools is a plus.
Have experience applying risk frameworks, threat models, severity assessments, taxonomies, or prioritization frameworks to structure current operational ambiguity and support decision-making.
Communicate complex topics clearly and effectively, producing concise, executive-ready analysis and influencing stakeholders across technical and non-technical teams.
Thrive in fast-moving environments, balancing rigor with pragmatism while building lightweight operating processes and tools that improve organizational coordination and follow-through on material risks.
About OpenAI
OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity. We push the boundaries of the capabilities of AI systems and seek to safely deploy them to the world through our products. AI is an extremely powerful tool that must be created with safety and human needs at its core, and to achieve our mission, we must encompass and value the many different perspectives, voices, and experiences that form the full spectrum of humanity.
We are an equal opportunity employer, and we do not discriminate on the basis of race, religion, color, national origin, sex, sexual orientation, age, veteran status, disability, genetic information, or other applicable legally protected characteristic.
For additional information, please see OpenAI’s Affirmative Action and Equal Employment Opportunity Policy Statement.
Background checks for applicants will be administered in accordance with applicable law, and qualified applicants with arrest or conviction records will be considered for employment consistent with those laws, including the San Francisco Fair Chance Ordinance, the Los Angeles County Fair Chance Ordinance for Employers, and the California Fair Chance Act, for US-based candidates. For unincorporated Los Angeles County workers: we reasonably believe that criminal history may have a direct, adverse and negative relationship with the following job duties, potentially resulting in the withdrawal of a conditional offer of employment: protect computer hardware entrusted to you from theft, loss or damage; return all computer hardware in your possession (including the data contained therein) upon termination of employment or end of assignment; and maintain the confidentiality of proprietary, confidential, and non-public information. In addition, job duties require access to secure and protected information technology systems and related data security obligations.
To notify OpenAI that you believe this job posting is non-compliant, please submit a report through this form. No response will be provided to inquiries unrelated to job posting compliance.
We are committed to providing reasonable accommodations to applicants with disabilities, and requests can be made via this link.
OpenAI Global Applicant Privacy Policy
At OpenAI, we believe artificial intelligence has the potential to help people solve immense global challenges, and we want the upside of AI to be widely shared. Join us in shaping the future of technology.