Technical Abuse Investigator

Stack

Responsibilities

• ▹Detect, investigate, and disrupt abuse and harm together with policy, legal, global affairs, security, and engineering teams via complex datasets
• ▹Develop and iterate on abuse signals and investigative methods, scaling one-off insights to reduce manual effort and expand coverage
• ▹Build and maintain lightweight technical solutions (e.g., SQL/Python data pipelines, investigation templates, dashboards, or internal utilities) for investigators focused on specific harm domains
• ▹Develop a deep understanding of OpenAI's products, data systems, and enforcement mechanisms, collaborating with engineering and data teams to improve tooling, data quality, and workflows
• ▹Communicate investigation findings effectively through written briefs, data-backed recommendations, and escalation summaries
• ▹Rotate periodically into an incident response role requiring rapid threat triage, investigation, mitigation, and concise leadership briefing
• ▹Participate in an on-call rotation to handle urgent escalations outside normal work hours

Requirements

• ▹Deep expertise in at least two of the following domains: agentic AI misuse, automation, encryption, terrorism, fraud, violence, child exploitation, data science, dashboarding, API abuse, product exploits, prompt injection, distillation
• ▹5+ years of experience investigating and mitigating abuse in a relevant domain
• ▹4+ years of relevant technical projects
• ▹Strong presenter on safety work in public or policy settings
• ▹Experience scaling or automating processes, especially with LLMs or ML techniques

Soft skills

About the company

The Intelligence and Investigations team supports OpenAI's mission by detecting, investigating, and disrupting misuse of its products, with a focus on critical or novel harms. Their work enables partner teams to build data-backed model policies and scalable safety mitigations.