Safety Manager in San Francisco

About the Team Safety Systems manages the complete lifecycle of safety efforts for OpenAI’s frontier models, ensuring our models are deployed responsibly and have a positive impact on society. Our work spans diverse research and engineering initiatives—from system-level safeguards and model training to evaluation and red-teaming—all aimed at mitigating misuse and maintaining our high bar for safety. We lead OpenAI's commitment to developing and deploying safe Artificial General Intelligence (AGI), fostering a culture of trust, responsibility, and transparency. Our goal is to continuously learn from deployments, distribute AI’s benefits widely, and ensure that powerful tools remain aligned with human values and safety considerations. About the Role The Safety Measurement Product Manager owns OpenAI's approach to measuring harm and safeguard efficacy in production, including driving the strategy for our suite of safety measurement platforms and products used across the company. You will partner closely with our safety research and engineering teams to determine what we measure, where we measure it, and how we measure it, feeding those insights directly into critical leadership decisions and back into our safety work. You will also represent the company's topline safety metric as well as prioritize incoming requests from partner teams to expand our safety measurement platform to more use cases. This position is based in San Francisco, CA, with relocation assistance available. In this role, you will: Partner closely with data science, research, engineering, policy teams, and other stakeholders to craft a vision for understanding safety outcomes and prevalence on our platforms. Define strategic priorities and product roadmaps focused on improving safety measurement approaches will scaling our measurement platform to more use cases, products, and cross-functional team needs. Establish repeatable processes to integrate cutting-edge AI safety research into OpenAI’s safety m

About the Team Safety Systems manages the complete lifecycle of safety efforts for OpenAI’s frontier models, ensuring our models are deployed responsibly and have a positive impact on society. Our work spans diverse research and engineering initiatives—from system-level safeguards and model training to evaluation and red-teaming—all aimed at mitigating misuse, misalignment, and maintaining our high bar for safety. We lead OpenAI's commitment to developing and deploying safe Artificial General Intelligence (AGI), fostering a culture of trust, responsibility, and transparency. Our goal is to continuously learn from deployments, distribute AI’s benefits widely, and ensure that powerful tools remain aligned with human values and safety considerations. Within Safety Systems, the Model Policy team works to ensure that frontier models behave safely and reliably in real-world environments by designing policies that define safe model behavior. Our relevant publications include: Safety at every step OpenAI GPT6 System Card OpenAI Model Spec GPT-Live ChatGPT Images 2.5 About the Role We are hiring a Model Policy Manager to focus on the safety of multimodal models. In this role, you will shape how OpenAI identifies, evaluates, and addresses risks in multimodal AI models - such as GPT-Live and ChatGPT Images - as well as multimodal capabilities in frontier AI models. This role is based in San Francisco, CA. We use a hybrid work model of 3 days in the office per week and offer relocation assistance to new employees. In this role, you will: Design and maintain model policies for audio, image, video, and omni-modal behavior. Translate theories of harm and threat models into behavioral safety policies, evaluation criteria, grading guidance, and safeguards. Identify and analyze safety regressions and failure patterns to identify gaps in existing policies and inform policy iteration. Develop policy artifacts that support model training, evaluation, and deployment, including behavior i

About the Team Our Safety Systems team is at the forefront of OpenAI's mission to build and deploy safe AGI, driving our commitment to AI safety and fostering a culture of trust and transparency. Within Safety Systems, the Model Policy team works to ensure that increasingly capable models behave safely and reliably in real-world environments. We investigate emerging model failures, define the behavior models should exhibit instead, and develop the data, evaluations, monitoring, and safeguards needed to improve and validate that behavior. Our work connects alignment research with the practical challenges of training and deploying frontier models. About the Role In this role, you will shape how OpenAI understands and addresses real-world risks that emerge from model misalignment as models become more autonomous and operate over longer horizons. You will investigate how misaligned behavior emerges across extended trajectories - including when models persist toward the wrong objective, take unsafe shortcuts, lose track of instructions, exploit weaknesses in their environment, or circumvent constraints - and translate these insights into behavioral policies, evaluations, monitoring, and safeguards. This role is ideal for someone who wants to turn alignment and safety concerns into concrete, empirically grounded improvements to frontier AI systems. Your Responsibilities: Identify vulnerabilities that emerge as models interact with tools, data, and external systems, and translate them into model- and system-level safeguards. Develop threat models and empirical frameworks for understanding harmful outcomes from misaligned behavior. Build frameworks for understanding harmful outcomes arising from model misalignment. Identify the underlying behaviors and system conditions that drive those outcomes. Turn findings into policy frameworks, evaluation criteria, online measurement and safeguards. Develop human data campaigns and gold sets to ground measurement and evaluation of eme

About the Team Safety Systems manages the complete lifecycle of safety efforts for OpenAI’s frontier models, ensuring our models are deployed responsibly and have a positive impact on society. Our work spans diverse research and engineering initiatives—from system-level safeguards and model training to evaluation and red-teaming—all aimed at mitigating misuse, misalignment, and maintaining our high bar for safety. We lead OpenAI's commitment to developing and deploying safe Artificial General Intelligence (AGI), fostering a culture of trust, responsibility, and transparency. Our goal is to continuously learn from deployments, distribute AI’s benefits widely, and ensure that powerful tools remain aligned with human values and safety considerations. About the Role We are hiring a Product Manager to focus on risk related to multimodal models. In this role, you will drive initiatives which ensure that OpenAI’s audio, image, and video deployments are safe, impactful, and aligned with user needs and technical innovation. You will clarify strategic priorities, develop safety-focused product roadmaps, and collaborate closely with AI researchers, software engineers, policy experts, and cross-functional partners. This role suits a proactive, technically skilled product manager adept at adversarial thinking and excited to tackle challenging, ambiguous problems through structured analysis and collaborative decision-making. This position is based in San Francisco, CA, with relocation assistance available. In this role, you will: Partner closely with AI research, engineering, data science, policy teams, and other stakeholders to embed safety throughout the development and deployment of multimodal AI models - such as GPT-Live and ChatGPT Images - as well as multimodal capabilities in frontier AI models. Develop comprehensive frameworks for understanding and mitigating deployment safety risks, drawing on data analysis, expert consultation, and adversarial assessments. Define strate

About the Team OpenAI’s mission is to ensure that artificial general intelligence benefits all of humanity. Safely delivering increasingly capable AI systems requires scalable technical safeguards, clear ownership of emerging risks, rigorous deployment readiness, and close coordination across research, engineering, product, operations, legal, policy, and external partners. Our Technical Program Managers lead complex, high-stakes initiatives that turn safety commitments into deployed systems and measurable outcomes. We work across model development, infrastructure, product, and operational response to help ensure our technology is deployed responsibly and cannot be used to cause serious real-world harm. About the Role We’re seeking Technical Program Managers to drive complex product, platform, and safety initiatives across ChatGPT, API, enterprise, and related deployment environments. These roles operate at the intersection of technical strategy and execution: you will turn safety and product priorities into actionable plans, influence architectural and operational decisions, and deliver durable capabilities across model, infrastructure, application, and platform layers. Depending on the role, you may enable sensitive or high-impact model deployments, integrate safeguards into cloud and API platforms, prevent violent misuse and other serious harms, improve detection and enforcement systems, create platform solutions for safety or establish new programs as risks evolve. You will partner deeply with engineers, researchers, product managers, and operational teams while communicating technical tradeoffs and program decisions to senior leadership. You bring technical fluency, product judgment, and a strong execution record. You’re comfortable navigating ambiguity, advocating for users and developers, balancing safety with model usefulness, and leading cross-functional work with urgency, rigor, and empathy. Specific focus areas and scope will vary by opening and level. Thi

About the Team OpenAI’s mission is to ensure the responsible and widespread adoption of artificial intelligence. In support of that mission, the Marketing team helps deeply understand customer audiences and market dynamics, influence the development of the right products, build sustainable and customer-aligned monetization models, and drive awareness, adoption, and usage across OpenAI’s products and platform. We take a data-driven approach to understand markets, develop monetization strategies, and uncover customer needs that shape product strategy and messaging. We partner closely with Sales, Partnerships, Product, Engineering, Research, Comms, and Design to deliver a cohesive end-to-end customer experience and lead go-to-market efforts for new product launches across channels. About the Role As a Product Marketing Manager, Research & Safety , you will shape how the world understands OpenAI’s frontier research and our approach to developing AI responsibly. You will establish the strategic narrative for some of the company’s most important research themes—including frontier capabilities, safety, preparedness, reasoning, and compute—and translate complex technical work into clear, credible stories for researchers, policymakers, and the broader public, working with cross-functional teams to translate that messaging and reach developers and enterprise. Working closely with Research, Safety Systems, Policy, Communications, and Product teams, you’ll lead the go-to-market strategy and messaging for major research and safety initiatives. You’ll develop narratives that build trust, translating technical concepts into tangible artifacts and messaging, drive industry understanding of emerging evaluation frameworks, and help communicate OpenAI’s unique perspective on responsible frontier AI development. You’ll also identify creative and non-traditional ways to bring these stories to the audiences that matter most. In this role, you will: Establish the strategic narrative f

About the Team At OpenAI, the User Safety & Risk Operations (USRO) team helps protect our products and users from abuse, fraud, safety risks, and other forms of misuse. We operate at the front line of real-world safety and risk management, translating user and operational signals into timely decisions, effective interventions, and improvements to our systems. This role sits on a team focused on building operational capacity for new, ambiguous, and fast-moving company priorities. The team defines what needs to be built, creates the operating model to support it, and works with partner teams to make the work scalable and durable over time. About the Role We are looking for a senior program manager to build the safety, quality, and risk operations supporting a new category of consumer devices. You will translate ambiguous product risks and evolving requirements into practical operating models, workflows, escalation paths, launch-readiness plans, and cross-functional decision-making. This is a foundational role: the systems you build will shape how OpenAI launches, monitors, and improves a new category of consumer devices safely at scale. You will help establish how potential safety incidents, product-quality concerns, sensitive customer escalations, privacy-sensitive issues, and other emerging device risks are identified, investigated, resolved, and incorporated into product and operational improvements. You will turn incomplete requirements into practical workflows, decision rights, launch plans, quality controls, measurement, and durable ownership. The role centers on program building, operational judgment, and execution. We welcome candidates from product safety, quality assurance, regulatory operations, technical program management, healthcare, medical devices, aerospace, consumer technology, and other environments involving complex products or regulated risks. Direct consumer-hardware experience is helpful but not required. The strongest candidates learn unfam

About the Team The Applied team brings OpenAI’s technology to the world through products used by hundreds of millions of people and by developers and businesses building on our APIs. We work across research, engineering, product, policy, safety, and operations to deploy frontier AI systems responsibly and safely. The Trust & Safety Data Engineering team builds the data foundations that help OpenAI understand, detect, investigate, and mitigate abuse and safety risks across our products. We partner with Integrity, Investigations, Safety Systems, Product Policy, Privacy, Data Science, Engineering, and Data Platform to create reliable, privacy-safe datasets and pipelines for fraud and abuse detection, enforcement workflows, safety measurement, ML feature generation, launch readiness, and transparency reporting. About the Role We are hiring a Technical Lead Manager to lead and grow the Trust & Safety Data Engineering team. This is a hands-on leadership role for someone who can set strategy, shape data architecture, align senior stakeholders, coach engineers, and drive execution on high-impact data systems. You will help turn fragmented launch and incident support into durable, reusable, privacy-safe data foundations that Trust & Safety teams can rely on. The systems your team builds will help OpenAI detect risk, investigate abuse, power operational workflows, develop and evaluate safety models, measure interventions, support product launches, and report accurately on platform integrity. In This Role, You Will Lead and grow a high-performing Trust & Safety Data Engineering team. Define the roadmap and technical strategy for Trust & Safety data systems. Build canonical, privacy-safe datasets and pipelines for abuse detection, fraud detection, risk signals, enforcement, scaled review, transparency reporting, and safety monitoring. Create reusable foundations for Trust & Safety model development, including features, labels, training data, backtesting,

About the Team The Frontier Assurance team brings independent scrutiny into OpenAI’s safety decisions and helps the public understand and assess our safety work. We lead third-party assessments and safeguard testing for OpenAI’s flagship launches, pilot new assurance mechanisms such as embedded auditing, run our misalignment disclosure process, and incorporate independent expert input as evidence for critical safety decisions. About the Role As a Research Program Manager on the Frontier Assurance team, you will build programs that bring independent expertise into frontier AI safety decisions and make the evidence behind those decisions understandable to the public. You will lead external research partnerships and third-party assessments, coordinate public safety documentation, and develop new approaches to independent scrutiny and transparency. Working across research, engineering, product, policy, and communications, you will help ensure external findings inform concrete decisions and that our public explanations accurately reflect the evidence, limitations, and remaining uncertainty. We’re looking for people with deep experience in research partnerships and program management with technical and research teams. This role combines partnership management, cross-functional coordination, an understanding of AI safety research, alignment, and evaluations, and strong communication skills. You will work with researchers and engineers within OpenAI and across the external community to initiate projects, set ambitious goals and milestones, and drive execution across multiple teams. This role is based in San Francisco, CA. We use a hybrid work model of 3 days in the office per week and offer relocation assistance to new employees. In this role, you will: Design and run third-party assessment programs for frontier models and safeguards, including independent evaluations, adversarial testing, and new approaches such as embedded auditing. Work with researchers and external part

About the Team Safety Systems manages the complete lifecycle of safety efforts for OpenAI’s frontier models, ensuring our models are deployed responsibly and have a positive impact on society. Our work spans diverse research and engineering initiatives—from system-level safeguards and model training to evaluation and red-teaming—all aimed at mitigating misuse, misalignment, and maintaining our high bar for safety. We lead OpenAI's commitment to developing and deploying safe Artificial General Intelligence (AGI), fostering a culture of trust, responsibility, and transparency. Our goal is to continuously learn from deployments, distribute AI’s benefits widely, and ensure that powerful tools remain aligned with human values and safety considerations. Within Safety Systems, the Model Policy team works to ensure that frontier models behave safely and reliably in real-world environments by designing policies that define safe model behavior. Some of our publications include: Safety at every step OpenAI GPT6 System Card OpenAI Model Spec About the Role We’re hiring a Model Policy Manager to shape model behavior for U.S. government use, with a focus on national security applications. You’ll define nuanced policies and translate them into training and evaluation criteria, helping models navigate high-stakes scenarios while preserving their usefulness and capabilities. In this role, you will: Develop model policies that guide safe and useful behavior. Build evaluations, identify policy gaps and model failures, and use findings to improve policies and training. Work with research, engineering, and domain experts to support safe, reliable deployment. You might thrive in this role if you: Bring relevant experience in AI safety, policy, or risk assessment. Have strong judgment and can turn complex safety questions into clear, practical policies. Have the technical fluency to work hands-on with model data and evaluations. Are motivated by OpenAI’s mission and the responsible use of

About the Team Critical Harm Operations sits within User Safety & Risk Operations and builds enforcement systems for Frontier Risk and Material Harm that are accurate, fast, defensible, and built to scale. We turn policy intent into operational readiness, review standards, quality systems, escalation paths, automation guardrails, and durable cross-functional operating models. About the Role We are looking for an exceptional Program Manager to help build durable operating systems and run some of OpenAI’s most complex safety operations. The core need is a high-agency operator who can take an ambiguous problem, create the right structure, align cross-functional partners, and drive the work through execution. This role will move across Critical Harm priorities as needs evolve. You may step into operationalizing national security or violent-activities workflows, support wellbeing and Frontier Risk initiatives, or help scale programs such as Trusted Access. Deep domain expertise is helpful but not required; the strongest candidates will learn quickly, exercise excellent judgment, and make complex programs move. In this role, you will: Lead strategic operational builds across priority workflows from problem statement to implemented operating model, including scope, owners, milestones, risks, success measures, and execution cadence. Translate policy, safety, technical, legal, and operational constraints into workflows, requirements, playbooks, escalation paths, and decision-making structures that teams can execute. Coordinate with User Ops leadership, Product Policy, Integrity, Safety Systems, i2, Legal, Product, Engineering, Support, vendors, and other partners to resolve dependencies and keep critical work moving. Move in and out of workflows as priorities shift—standing up new programs, stabilizing operations, improving handoffs, and transitioning durable ownership to the right team. Use operational data and frontline signals to identify bottlenecks, quality gaps, ca

About the Team OpenAI's Environmental, Health & Safety (EHS) team partners across the company to enable safe, responsible growth. This role will be a senior EHS partner to our robotics operations as our operating footprint expands. You will work closely with Robotics, Engineering, Operations, Facilities, Workplace, Construction, Security, People, Legal, and external partners to integrate safety into how our facilities are designed, built, staffed, and operated. About the Role We are seeking a Senior Manager, EHS - Robotics to provide senior, hands-on safety leadership across multiple sites. This role is for an experienced EHS leader who can operate independently in a fast-moving technical environment, anticipate risk before it becomes an incident, and build practical safety systems that scale with the business. Our robotics operations are scaling quickly, with a growing mix of construction, commissioning, workforce expansion, and around-the-clock operations. This creates a complex and evolving risk environment and requires strong preventive planning, consistent field presence, and sustained incident-management oversight. This leader will own site-level EHS strategy and execution for the robotics portfolio while partnering with the broader EHS team on company-wide standards and programs. The successful candidate will look around corners: planning for future ramps, identifying requirements early, influencing design and operating decisions, and creating durable systems that enable teams to move quickly without compromising safety. In this role you will: Serve as a senior EHS partner across robotics R&D operations, owning site-level safety strategy, priorities, and execution as the footprint scales. Anticipate EHS requirements for new operations, equipment, processes, construction phases, staffing ramps, and 24/7 operations, and translate them into practical plans before work begins. Lead proactive risk identification and control across robotics activities, incl

About the Team The Statsig team is responsible for the experimentation, feature rollout, dynamic configuration, and analytics systems that help OpenAI ship products with speed, safety, and evidence. Teams across ChatGPT, Codex, model measurement, monetization, business subscriptions, developer products, and shared infrastructure rely on Statsig to introduce capabilities safely, measure their impact, and make high-confidence product decisions. About the Role As a Product Lead on the Statsig team, you will define how experimentation, rollout, configuration, and analytics become a simple, reliable, and trusted part of how every OpenAI product team ships. You will set strategy across multiple product and platform workstreams, translate company-wide needs into durable capabilities, and help Statsig become a core part of OpenAI’s product development system. We’re looking for a product leader who combines strong product judgment, technical fluency, and deep analytical thinking. You should be comfortable navigating ambiguous customer needs, influencing teams across the company, and balancing rapid adoption with reliability, usability, and measurement quality. This role is based in San Francisco, CA. We use a hybrid work model of 3 days in the office per week and offer relocation assistance to new employees. Travel requirements should be confirmed with the recruiter before publishing. In this role, you will: Define the product vision, strategy, and roadmap for experimentation, feature management, dynamic configuration, rollout safety, and analytics. Partner with product, engineering, research, data, design, and infrastructure leaders to turn recurring launch and measurement needs into reusable platform capabilities. Develop a deep understanding of workflows across ChatGPT, Codex, model measurement, monetization, subscriptions, and developer products, then establish clear priorities across competing needs. Drive adoption by making sophisticated experimentation and analytics c

About the Team OpenAI's Research Data Team exists to accelerate the evaluation, safety and capabilities of our models and products. Made up of technical operators and software engineers, we design the methods in which we acquire and create data. About the Role As a Research Program Manager (RPM), Data Acquisition, you will partner with research, engineering, and operations to design and implement pragmatic solutions for acquiring data. You will be a key interface between our research roadmap and external data offerings. This role is based in our San Francisco HQ and will be part of a team of RPMs pushing the frontier of data acquisition. In this role, you will: Partner deeply with research: Work with researchers to scope data needs, define success criteria, and translate priorities into clear execution plans. Shape the data acquisition pipeline: Identify, evaluate, and advance high impact data opportunities - balancing research value, feasibility, quality, and responsible execution. Unblock yourself: Move work forward even when the path is unclear — using technical judgement, creative problem solving, and scrappy execution to make progress while longer-term solutions are still forming. Build lightweight systems and visibility: Use SQL, Python, dashboards, and simple tooling to track performance, quality, and blockers. Drive technical roadmaps: Collaborate with engineers to enhance data platforms, resolve blockers, and ensure security best practices such as access management. Scale your impact: Equip vendors and internal teams with the context, standards, and operating rhythms needed to focus on the most important problems. You’ll thrive in this role if you: Are proficient in SQL and Python for analysing datasets, querying databases, building dashboards, and generating actionable insights. Are comfortable using APIs, automation, and AI tools such as Codex to accelerate workflows, remove manual overhead, and upskill quickly in unfamiliar technical areas.Experience sou

About the Team The Integrity team builds the systems OpenAI uses to understand, prevent, and respond to misuse across our products. We partner with Product, Policy, Safety Systems, User Operations, Security, Legal, Privacy, OpenAI for Government, and research teams to turn policy and threat models into product controls, review workflows, measurement systems, and enforcement paths. About the Role We are hiring a Product Manager to own Integrity's product strategy for sensitive deployments: contexts where model capability, customer or deployment setting, privacy constraints, and misuse potential make the operating bar unusually high. This includes government and other high stakes deployments, regulated or high-trust enterprise contexts, zero data retention and privacy-constrained environments, and agentic workflows where harm can unfold across many steps rather than a single prompt. The Sensitive Deployments PM will focus on high-consequence use cases relevant to government deployments and broader deployment-readiness questions for high-risk domains (e.g., healthcare), while building reusable Integrity capabilities for agentic detection and enforcements in these environments. This position is based in San Francisco, CA, with relocation assistance available. In this role, you will: Own the roadmap for sensitive deployment Integrity controls and readiness criteria. Define how we measure residual risk, decision quality, review quality, and mitigation effectiveness. Build agentic investigation and context-assembly workflows for complex, multi-step misuse patterns. Partner with Policy, Legal, Privacy, Safety Systems, User Ops, Government, and product teams to build shared capabilities. Create launch and deployment playbooks for sensitive customer, capability, and product contexts. You might thrive in this role if you: Have shipped complex product systems in AI, integrity, trust and safety, security, privacy, risk, government, regulated enterprise, or AI safety. Can turn am