Threat Intel Manager, Model Exploitation & Fraud

Written by the T&S Network from Anthropic's posting.

Builds and runs a Model Exploitation & Fraud team inside Threat Intelligence that detects, investigates, and disrupts large-scale exploitation of an AI company's models. Covered abuse includes model distillation, unauthorized access, account farming and reseller abuse, and fraud and scam operations. The team's job is to direct and resource senior investigators tracing distillation networks, reseller and proxy ecosystems, and financially motivated actors across first-party and third-party platforms.

• Fraud and scams
• AI misuse and jailbreaks

What you'd do

• Set strategy, priorities, and the detect, investigate, action, and share scope for the mission area
• Hire, manage, and develop a small team of technical threat investigators and set the casework quality bar
• Define lanes between management and senior individual contributors who own the deepest technical investigations
• Steer and staff complex probes into distillation, account misuse, unauthorized access, and fraud rings
• Redesign triage for a high-volume detection pipeline using abuse signals, clustering, and agentic workflows
• Create fraud and scam detection and investigation playbooks starting from nothing
• Lead the external sharing program with U.S. government agencies and industry peers

What it asks for

• Proficiency in SQL and Python for reviewing data-heavy casework
• Experience overseeing investigations across surface, deep, and dark web, including reseller and access-broker communities
• Record of building processes, detection systems, or programs from nothing and showing what changed
• Clear communication with executives, engineers, and external partners

Worth knowing

• Exposure to explicit sexual, violent, or psychologically disturbing content
• Escalation responses required during weekends and holidays

Our best summary of the posting, and it can be wrong. The employer's posting is what counts: read it in full on Anthropicnorth_east before you apply. If we got something wrong, report this listing.