The work
You will assess whether AI systems reason correctly about defense, international law, military support, and lawful policy boundaries. The work includes difficult cases involving the law of armed conflict, rules of engagement, operational advice, and legal uncertainty.
You will provide clear legal analysis in writing and conversation, explain disputed cases, and help improve how AI models handle legitimate legal and policy questions.
- Create scenarios that distinguish lawful and unlawful military policy review and support.
- Check category boundaries using international humanitarian law and the law of armed conflict.
- Review and grade AI responses for legal reasoning and appropriate treatment of uncertainty.
- Identify cases where a model incorrectly refuses a legitimate policy or legal question.
- Work with other subject-matter experts to reach sound conclusions on disputed cases.
What it pays and takes
This is remote, part-time contractor work.
- Pay: $70 to $120 per hour.
- Language: Professional fluency in English.
- Location: Open to candidates in the countries listed for this role, including the United States, Canada, Brazil, India, Japan, Singapore, South Africa, and other approved countries.
- Experience: A Judge Advocate background or substantial experience in international humanitarian law and the law of armed conflict.
- Legal knowledge: Fluency in rules of engagement and lawful boundaries affecting military or surveillance operations.
- Professional background: Experience advising on operational legal matters in military or government settings.
- Core skills: Strong written and verbal communication, policy review, fact-pattern analysis, and sound reasoning about legal uncertainty.
- Helpful background: Teaching, consulting, or advisory work in military or international law, plus familiarity with AI governance, ethics, or legal frameworks for AI.
How it works
You apply here in about two minutes and finish on the hiring platform. The next step is a short AI interview (technical roles include a coding challenge). Passing it is how you qualify for paid tasks, so take it soon after you are invited.
The hiring platform's application for this job asks for your resume, your LinkedIn profile, your phone number, your location and answers to 5 screening questions.
About AI training work
AI training is the human work behind systems that generate and evaluate text, images, code, and other content. People with specialized knowledge review examples, rate model responses, and explain errors so AI systems handle complex subjects more accurately and responsibly. Experienced contributors are paid for applying that judgment to challenging cases.