We are looking for highly experienced individuals with backgrounds in intelligence, defense, cyber operations, counterintelligence, counterterrorism, and related fields to join our AI research community.
Your role will be to simulate and evaluate how malicious actors might leverage large language models (LLMs) to plan and enhance acts of violence, terrorism, targeted assassinations, special operations, or tactical-level attacks. Fellows will engage in red teaming exercises, evaluating and interpreting LLM-generated outputs from jailbreaking attempts, and determining whether those outputs constitute credible, material threats in real-world scenarios.
This is a high-impact position ideal for professionals who want to apply their government or military background to anticipate national security risks in the age of AI.
Overview:
Who Should Apply:
We are seeking seasoned professionals with firsthand operational, analytic, or planning experience in the national security, defense, or intelligence communities. Candidates should be comfortable reviewing sensitive or graphic content for security relevance and applying real-world judgment in ambiguous scenarios.
You likely have experience such as:
Preferred Background and Education:
While experience is paramount, we also welcome candidates with advanced academic training from elite defense or public policy institutions.
Good Candidates Come From:
Why Join Us:
Apply Today to Be Considered for Upcoming Opportunities:
Submitting your application now ensures you’re added to our candidate pool for upcoming projects. You may be matched with a project shortly after applying, or it may take some time depending on availability. In either case, applying today puts you one step closer to future opportunities as they arise.
We are seeking seasoned professionals with firsthand operational, analytic, or planning experience in the national security, defense, or intelligence communities. Candidates should be comfortable reviewing sensitive or graphic content for security relevance and applying real-world judgment in ambiguous scenarios.
F-1 students who are eligible for CPT or OPT may be eligible for projects on Handshake AI. Work with your Designated School Official to determine your eligibility. If your school requires a CPT course, Handshake AI may not meet your school’s requirements. STEM OPT is not supported. See our Help Center article for more information on what types of work authorizations are supported on Handshake AI.