AI Safety Expert – English, Danish
Employment Information
Report this job
Job expired or something wrong with this job?
Job Description
AI safety red team expert probing Mercor’s frontier AI models for vulnerabilities. Generating reproducible attack data that strengthens customer AI safety and robustness.
Responsibilities:
- Red-team conversational AI models and agents through jailbreaks, prompt injections, misuse cases, bias exploitation, and multi-turn manipulation
- Generate human data by annotating failures, classifying vulnerabilities, and flagging systemic risks
- Apply taxonomies, benchmarks, and playbooks to keep testing consistent
- Produce reproducible reports, datasets, and attack cases for customers
- Identify vulnerabilities that automated tests miss
- Expand evaluation coverage and reduce production surprises
- Help customers strengthen the safety, robustness, and trustworthiness of AI systems
- Work on projects training and enhancing frontier AI systems
Requirements:
- Native fluency in English and Danish is required
- Prior red teaming experience in AI adversarial work, cybersecurity, or socio-technical probing
- Ability to probe AI systems adversarially and push systems to breaking points
- Experience using frameworks or benchmarks for structured testing
- Ability to explain risks clearly to technical and non-technical stakeholders
- Ability to adapt across projects and customers
- Nice-to-have: adversarial ML experience, including jailbreak datasets, prompt injection, RLHF/DPO attacks, or model extraction
- Nice-to-have: cybersecurity experience, including penetration testing, exploit development, or reverse engineering
- Nice-to-have: socio-technical risk experience, including harassment/disinformation probing, abuse analysis, or conversational AI testing
- Nice-to-have: psychology, acting, or writing experience for unconventional adversarial thinking
- Must be an independent contractor
- Candidates must not require H1-B or STEM OPT support
Benefits:
- Fully remote role
- Flexible schedule; work can be completed on your own schedule
- Weekly payments via Stripe or Wise based on services rendered
- Projects may be extended, shortened, or concluded early depending on needs and performance
- Participation in higher-sensitivity projects is optional
- Clear guidelines and wellness resources for sensitive-content work
- Competitive pay
- Referral payments of up to $250 for each successful referral
Similar Jobs
AI Safety Red Teamer
AI Safety Red Teamer stress-testing frontier AI models for Mercor. Identifying vulnerabilities and improving model alignment through adversarial evaluation.
AI Safety Red Teamer
AI Safety Red Teamer adversarially testing frontier AI systems for Mercor's AI lab partners. Identifying jailbreaks, hallucinations, policy failures, and safety vulnerabilities.
AI Ops Engineer
AI Operations Engineer managing enterprise AI platforms for Gruve, an AI-focused software services startup. Supporting users, optimizing licenses, and building agents and integrations.
AI Value Engineer
AI Value Engineer helping PhoenixTeam’s mortgage clients adopt generative AI workflows. Leading experiments, client enablement, and GenAI learning labs for regulated environments.
AI Safety Red Teamer
AI Safety Red Teamer adversarially testing Mercor’s frontier AI systems. Identifying jailbreaks, hallucinations, and policy failures across high-risk domains.
AI Safety Red Teamer
AI Safety Red Teamer uncovering vulnerabilities in frontier AI models for Mercor. Designing adversarial tests and improving model safety with leading researchers.
AI Enablement Engineer
AI Enablement Engineer building AI tools, automations, and production systems for Phorest’s global salon software platform. Measuring business impact and scaling effective solutions.
AI Quality Evaluator, Italian
AI Quality Evaluator assessing Italian-language AI content, datasets, and designs for Canva’s global design platform. Ensuring linguistic quality, accuracy, and cultural relevance.
AI Quality Evaluator – Spanish
AI Quality Evaluator assessing Spanish AI-generated design content for Canva. Creating datasets and labels to improve localisation quality for Spanish users.
AI Facial Data Collection Associate, Twins
Twin participants submitting unedited facial photos and videos for 24-MAG’s AI biometric data research project. Completing separate applications and quality-controlled digital submissions.
AI Safety Experts, English – Punjabi
AI safety red teamer probing conversational models for Mercor’s human-data AI projects. Generating adversarial datasets, vulnerability classifications, and reproducible attack reports.
Working Student – Innovation, AI & Entrepreneurship
Werkstudent für AI- und Innovationsprogramme bei G+D, einem globalen Sicherheitstechnologieunternehmen. Analyse von Geschäftsideen, Entwicklung von Innovation Communities und Erprobung neuer AI-Tools.
AI Solution Consultant – AI Transformation Consultant
AI Solution Consultant bei TP, Weltmarktführer im digitalen Business Process Outsourcing. Entwicklung kundenorientierter AI-Lösungen, Roadmaps und Business Cases für bessere Customer Experience.
AI Tutor – Video, Weekend
Video Specialist training SpaceXAI’s Grok AI to interpret and generate video content. Producing annotated, curated video data using editing, motion graphics, and VFX expertise.
AI Safety Red Teamer
AI Safety Red Teamer stress-testing frontier AI models for Mercor. Identifying jailbreaks, hallucinations, and policy failures to improve AI safety.
AI Safety Experts – English, Norwegian
AI safety red team expert probing Mercor’s partner AI models for vulnerabilities. Generating adversarial datasets, attack cases, and reproducible safety reports.
AI Safety Expert, English, Finnish
AI safety red team expert probing Mercor’s frontier AI models for vulnerabilities. Generating adversarial datasets, reports, and attack cases to improve customer AI safety.
AI Safety Expert – English, Norwegian
AI safety red team expert probing Mercor’s partner AI models for vulnerabilities. Generating adversarial datasets, attack cases, and reproducible safety reports.
AI Safety Red Teamer
AI Safety Red Teamer stress-testing frontier models for Mercor, which partners with AI labs to train and enhance AI systems. Identifying vulnerabilities across high-risk domains.
AI Safety Red Teamer
AI Safety Red Teamer identifying vulnerabilities in frontier AI systems for Mercor’s AI training projects. Designing adversarial prompts and evaluating high-risk model behaviors.