AI Safety Red-Teamer — English & Danish Fluency Required

About the role

We're building a specialized red-teaming bounty for bilingual experts fluent in **English and Danish** to help stress-test AI systems before they reach the real world. The core belief driving this work: the safest AI is the one that has already been attacked — by us.

You'll join a distributed team of human evaluators who probe AI models with adversarial prompts, uncover weaknesses, and generate high-quality red-team data that directly improves model safety. This work touches sensitive and nuanced topics — including bias, misinformation, and harmful behavior patterns — so careful judgment, cultural fluency, and native-level command of both languages are essential.

This is a flexible, fully remote bounty ideal for linguists, safety researchers, trust & safety professionals, or subject-matter experts who want to contribute meaningfully to responsible AI development.

What you'll do

• Craft adversarial prompts designed to surface model vulnerabilities, biases, and unsafe outputs
• Evaluate AI-generated responses across English and Danish for accuracy, safety, tone, and cultural nuance
• Document findings clearly, categorizing failure modes and severity
• Collaborate with a distributed red-team cohort to refine testing methodologies
• Review sensitive content responsibly, following strict safety and confidentiality protocols
• Provide structured feedback that feeds directly into model retraining and safety improvements

Requirements

• Native or near-native fluency in **both English and Danish** (written and spoken)
• Strong critical thinking skills and comfort engaging with sensitive or controversial content
• Prior exposure to content moderation, linguistics, AI evaluation, trust & safety, or research work is a plus
• Excellent written communication and attention to detail
• Ability to work independently and meet deadlines in a fully remote setting
• Comfortable navigating ambiguous or ethically nuanced material with discretion and professionalism

Compensation

This bounty pays **$48–$62/hour**, based on experience and evaluation performance. Fully remote, flexible scheduling, with opportunities for ongoing work as the red-teaming program expands.