AI Safety Expert — English & Norwegian

📋 Description

• Remote, contract opportunity focusing on red-team safety for conversational AI.
• Test jailbreaks, prompt injections, misuse cases, bias exploitation, and multi-turn manipulation.
• Generate high-quality human evaluation data by annotating model failures and vulnerabilities.
• Apply frameworks and playbooks to ensure evaluations are consistent and reproducible.
• Create structured attack cases and reports with actionable insights for AI safety improvements.
• Develop adversarial approaches to uncover unexpected failure modes.

🎯 Requirements

• Native-level fluency in both English and Norwegian, excellent writing in both.
• Prior experience in red teaming, AI adversarial testing, cybersecurity, or related fields.
• Strong curiosity and adversarial mindset; ability to systematically push AI systems to limits.
• Experience with structured frameworks, benchmarks, taxonomies, or testing methodologies.
• Strong analytical and documentation skills; ability to explain vulnerabilities and risks clearly.
• Ability to work independently and adapt to changing project requirements.

🎁 Benefits

• Fully remote work with flexible scheduling.
• Independent contractor engagement with project-based flexibility.
• Weekly payments through Stripe or Wise.
• Exposure to human-data-driven AI safety and red teaming projects.
• Direct contribution to more robust, safe, trustworthy AI systems.
• Experience with cutting-edge AI safety projects and evolving evaluation methodologies.