Only AI Jobs


AI Emerging Risks Analyst

ID: 9117

Type: Full-time

Category: Others

Company Name: Rockset

Location: California (USA)

Education Level: Senior (5-10 years)

Visit company vacancy
Job Description

About the team

The Intelligence and Investigations team seeks to rapidly identify and mitigate abuse and strategic risks to ensure a safe online ecosystem in close collaboration with our internal and external partners. 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.

About the role

We are looking for an AI emerging risks analyst to help us understand potential harms and misuse of AI at the frontier in a time of rapid, sustained change. From known threat actors misusing new technologies to new threats enabled by new technologies, we seek to scan available signals and use strategic foresight methodologies to enable proactive detection and mitigation of frontier AI risks.

In this role, you will help to provide strategic-level perspective on a range of evolving risk areas, helping to produce actionable understanding of frontier AI risks relevant to OpenAI’s platforms, surfaces, and broader business interests. Utilizing mixed quantitative and qualitative methodologies, you will spot early warning signs, pull threads on potentially concerning behavior, and turn weak signals into clear, prioritized risk calls. You will focus on upstream ecosystem scanning, competitive benchmarking, and external narrative/risk sense-making. Your work will help to inform cross-functional partners in the protection and safety stacks to guide mitigations that keep users, brands, and communities safe while allowing productive, creative uses of these tools to thrive.

In this role, you will

  • Map and prioritize emerging risks at the frontier of AI

    • Build and continuously refine a clear picture of emerging signals and trends that could affect the AI ecosystem through upstream and external scanning.

    • Design and maintain harm taxonomies that provide foresight and warning about how AI harms and misuse may manifest over the next 0-24 months and beyond.

    • Contribute to an evergreen frontier risk register and prioritization framework that surfaces the top issues by severity, prevalence, exposure, and trajectory.

  • Detect and deep dive into emerging abuse patterns

    • Create comprehensive approaches to horizon scanning, competitive benchmarking, and external narrative/risk sense-making.

    • Stay current on abuse trends ranging from state actor misuse to criminal activity, drawing from the work of internal organizational and cross-functional partners.

    • Connect individual incidents into system-level stories about actors, incentives, product design weaknesses, and cross-product spillover–whenever possible spotting these incidents or even hypothesizing them before they hit our surfaces.

  • Turn analysis into actionable risk intelligence

    • Translate findings into clear, ranked risk lists and concrete proposals for mitigations that product, safety, and policy teams can execute on.

    • Work with Global Affairs and Communications teams to share findings in ways that reinforce OpenAI’s role as a leader in the online safety ecosystem.

    • Track whether mitigation work is landing: follow key indicators, pressure-test assumptions, and push for course corrections when the data demands it.

  • Build early warning and measurement capabilities

    • Help define the core metrics and signals that indicate whether fast-evolving AI environments are safe (e.g., key harm prevalence, severity distributions, escalation rates, brand safety issues).

    • Work with data science and visualization colleagues to shape monitoring views and dashboards that highlight leading indicators and unusual changes from signals spotted off platform to determine whether these are manifesting in user behavior or abuse patterns.

    • Pioneer new uses of our own technologies to scale detection and transform workflows.

  • Provide strategic analysis and future-looking perspectives

    • Produce concise but comprehensive strategic intelligence estimates that provide full context about a given interest area that includes confidence levels based on observed data to inform judgments and recommendations.

    • Run scenario analyses that explore how AI harms might evolve over the next 6–24 months (e.g., how scams may fundamentally evolve alongside the proliferation of agentic AI; how state actors may seek to misuse new scientific capabilities of frontier models).

    • Help design and run tabletop exercises for internal and partner audiences that distill manifest and latent risks at the frontier of AI and identify mitigations.

    • Benchmark OpenAI’s risk profile and mitigations against external incidents and other platforms, highlighting gaps, strengths, and opportunities.

  • Shape safety readiness for new products

    • Contribute to product readiness and launch reviews by laying out expected abuse modes based on broad, upstream understanding.

    • Turn risk insights into practical guidance for internal teams (product, marketing, partnerships, comms) and, where appropriate, external partners using OpenAI technologies in social and brand contexts.

    • Develop reusable frameworks, playbooks, FAQs, and briefing materials that make it easier for the broader organization to understand AI risks and respond consistently.

You might thrive in this role if you

  • Significant experience (typically 5+ years) in trust and safety, integrity, security, policy analysis, or intelligence work focused on a range of emerging risks situated in strategic context and translated into actionable intelligence.

  • Demonstrated ability to analyze complex online harms (e.g., harassment, coordinated abuse, scams, influence operations, brand safety issues) and convert all-source analysis into concrete, prioritized recommendations.

  • Strong analytical skills and comfort working with both qualitative and quantitative inputs, including: (1) Casework, incident reports, OSINT, product context, and policy frameworks. (2) Basic metrics and trends in partnership with data science (e.g., harm prevalence, severity profiles, exposure, escalation rates).

  • Strong adversarial and product intuition, able to foresee how actors might adapt AI tools for misuse and evaluate how product mechanics, incentives, and UX decisions influence risk.

  • Experience designing and using risk frameworks and taxonomies (e.g., harm classification schemes, severity/likelihood matrices, prioritization models) to structure ambiguous spaces and support decision-making.

  • Understanding of the application of foresight methodologies including horizon scanning, scenario planning, tabletop exercises, or simulations.

  • Proven ability to work cross-functionally with product, engineering, data science, operations, legal, and policy teams, including pushing for clarity on tradeoffs and following through on mitigation work.

  • Excellent written and verbal communication skills, including experience producing concise, executive-ready briefs and explaining sensitive, complex issues in grounded, concrete terms.

  • Comfort operating in fast-changing, ambiguous environments: you can identify weak signals, form hypotheses, test them quickly, and adjust as the product and threat landscape evolves.

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.

Company Information

Company Name: Rockset

Company Website: https://rockset.com

Company Address: San Mateo, California, USA

Rockset is a cloud-native data platform company that builds and operates a managed, real-time analytics database designed to enable fast, low-latency SQL queries on continuously changing data streams and operational data stores. The company positions its core product as a purpose-built service for application developers, analytics engineers, and data teams who need to run complex queries—including joins, aggregations, and full-text search—against high-volume, high-velocity data without managing database infrastructure. Rockset’s offering emphasizes instant ingestion, automatic indexing, and serverless scalability so customers can get real-time insights from event streams, operational databases, and data lakes with minimal operational overhead. At the center of Rockset’s business activities is a managed database service delivered from the cloud. The service is designed to ingest data from a variety of sources—change data capture (CDC) streams, message queues, object storage, and direct connectors to operational databases—and make that data queryable immediately. Rockset automates the work that traditional analytics stacks require: it handles continuous ingestion, schema evolution, indexing, resource provisioning, and query optimization. The company provides a hosted, fully managed experience that abstracts the underlying infrastructure and offers predictable developer ergonomics and APIs. Rockset’s main product is the Rockset cloud database service, a serverless, managed analytics engine that supports ANSI SQL and enables low-latency queries on fresh data. Key product capabilities include continuous data ingestion from sources such as streaming platforms and change streams, automatic and adaptive indexing (often referred to as a “converged index” approach combining row, columnar, and inverted-index elements), a SQL query interface for expressive analytics and search, and a pay-for-what-you-use consumption model. Because of the indexing approach and the system’s ability to shard and parallelize queries across cloud infrastructure, Rockset aims to deliver millisecond-to-subsecond query latencies even on complex queries over large, rapidly changing datasets. Integration and extensibility are important parts of Rockset’s product strategy. The platform provides connectors and ingestion paths for common cloud and data stack components, enabling users to stream or batch data from sources such as message brokers, operational databases (via CDC), cloud object storage, and data warehouses. On the query and consumption side, Rockset supports standard interfaces that allow integration with applications and analytics tools: developers can query with SQL through REST APIs, client libraries, and standard connectors so that Rockset can be used as a back-end for dashboards, embedded analytics, personalization, recommendation engines, and search-oriented application features. Rockset also provides capabilities to join and enrich streaming or operational data with historical datasets, enabling hybrid analytical workloads that combine fresh and reference data. Operational characteristics of Rockset’s managed service emphasize elasticity and minimal operational burden. The platform is designed to be serverless from the user’s perspective—resources scale automatically to meet query and ingestion demand, and the user pays for compute and storage consumption rather than provisioning fixed clusters. Rockset also focuses on handling schema changes automatically as incoming data evolves, reducing the need for manual schema migrations that are common in traditional data warehouses. The managed nature of the service includes monitoring, automated failure recovery, and tuning handled by the platform, allowing engineering teams to focus on application logic and analytics rather than database administration. From a usage perspective, Rockset targets a set of real-time and near-real-time analytics use cases where freshness, performance, and developer productivity are critical. Typical applications include real-time dashboards and observability, customer 360 and personalization use cases, operational analytics, search-and-query experiences embedded in user-facing applications, and ad-hoc analytics on streaming event data. Rockset’s SQL-first approach helps lower the barrier to entry for analytics and data engineering teams already familiar with relational query language while also serving developers who need to embed queries in production applications. Security, governance, and enterprise readiness are presented as part of the managed offering: Rockset provides typical enterprise controls such as authentication and authorization, support for private network connectivity to cloud resources, and features intended to meet organizational security policies. The platform also exposes monitoring and auditing capabilities that help teams observe query performance and data flows. Rockset’s positioning is that customers can get the agility of real-time analytics while retaining operational safeguards required by corporate and regulatory environments. The company’s go-to-market focus combines direct product-led adoption—where developers can sign up and begin ingesting data and running queries quickly—with enterprise engagement for larger or production deployments that require integration, security reviews, and architectural guidance. Rockset’s documentation, SDKs, and examples emphasize rapid onboarding and practical recipes for building analytics and search features on top of streaming and operational data. Overall, Rockset is widely characterized in public materials as a cloud-native, real-time analytics database that abstracts infrastructure complexity and provides a SQL interface for immediate querying of continuously updated data. The product is aimed at teams that need real-time insights and application-facing analytics without the cost and complexity of managing specialized infrastructure or building custom indexing and ingestion pipelines.
Visit company vacancy