Intelligence Analyst for AI Red Teaming

Intelligence experts evaluate how malicious actors could use large language models to plan or enhance violent acts, terrorism, targeted assassinations, special operations, or tactical-level attacks. In this part-time, remote AI trainer role you will run adversary simulations and red team exercises to assess whether LLM outputs present credible, material threats in real-world contexts.

• Simulate how malicious actors might leverage LLMs to plan or improve acts of violence, terrorism, targeted assassinations, special operations, or tactical-level attacks.
• Conduct red teaming and jailbreaking exercises against LLMs, generating and testing prompts and adversary workflows.
• Evaluate and interpret LLM-generated outputs from jailbreaks and red team tests.
• Determine whether specific outputs constitute credible, material threats when mapped to real-world scenarios.

• Background in intelligence, defense, cyber operations, counterintelligence, counterterrorism, or closely related fields.
• Bachelor''s, Master''s, or PhD degree, or equivalent professional experience.
• Proven experience assessing operational threats, adversary tradecraft, or tactical planning is strongly preferred.

• Location: Remote.
• Employment type: Part-time.

• Up to $200/hr (depending on the project)

• Open to experienced professionals with government, military, or equivalent operational backgrounds who can perform remote, part-time work.

This position is well suited to candidates who want to apply government or military experience to anticipating national security risks from advanced AI. When applying, highlight your relevant intelligence, defense, cyber, counterintelligence, or counterterrorism experience and your highest degree earned. Applications should clearly describe past work that demonstrates the ability to assess adversary intent and operational feasibility in real-world scenarios.