adversarial ai tester

YO IT CONSULTING

<div>Work Mode: Remote 
Engagement Type: Independent Contractor 
Schedule: Full-Time or Part-Time Contract
Fluent Language Skills Required: English &amp; Spanish. Native-level fluency in English and Spanish is required for this position.
Role:
We believe the safest AI is the one that&rsquo;s already been attacked &mdash; by us. We are assembling a red team for this project - human data experts who probe AI models with adversarial inputs, surface vulnerabilities, and generate the red team data that makes AI safer for our customers.
This project involves reviewing AI outputs that touch on sensitive topics such as bias, misinformation, or harmful behaviors. All work is text-based, and participation in higher-sensitivity projects is optional and supported by clear guidelines and wellness resources. Before being exposed to any content, the topics will be clearly communicated.
What You&rsquo;ll Do
<ul><li>Red team conversational AI models and agents: jailbreaks, prompt injections, misuse cases, bias exploitation, multi-turn manipulation
</li><li>Generate high-quality human data: annotate failures, classify vulnerabilities, and flag systemic risks
</li><li>Apply structure: follow taxonomies, benchmarks, and playbooks to keep testing consistent
</li><li>Document reproducibly: produce reports, datasets, and attack cases customers can act on
</li></ul>Who You Are
<ul><li>You bring prior red teaming experience (AI adversarial work, cybersecurity, socio-technical probing)
</li><li>You&rsquo;re curious and adversarial: you instinctively push systems to breaking points
</li><li>You&rsquo;re structured: you use frameworks or benchmarks, not just random hacks
</li><li>You&rsquo;re communicative: you explain risks clearly to technical and non-technical stakeholders
</li><li>You&rsquo;re adaptable: thrive on moving across projects and customers
</li></ul>Nice-to-Have Specialties
<ul><li>Adversarial ML: jailbreak datasets, prompt injection, RLHF/DPO attacks, model extraction
</li><li>Cybersecurity: penetration testing, exploit development, reverse engineering
</li><li>Socio-technical risk: harassment/disinfo probing, abuse analysis, conversational AI testing
</li><li>Creative probing: psychology, acting, writing for unconventional adversarial thinking
</li></ul>What Success Looks Like
<ul><li>You uncover vulnerabilities automated tests miss
</li><li>You deliver reproducible artifacts that strengthen customer AI systems
</li><li>Evaluation coverage expands: more scenarios tested, fewer surprises in production
</li></ul><h2 class="jobSectionHeader">Contract and Payment Terms</h2>
<ul><li>You will be engaged as an independent contractor.</li>
<li>This is a fully remote role that can be completed on your own schedule.</li>
<li>Projects can be extended, shortened, or concluded early depending on needs and performance.</li>
<li>Payments are weekly on Stripe or Wise based on services rendered.</li></ul></div>

Spanish Ai Trainer - Remote

<div>Work Mode: Remote 
Engagement Type: Independent Contractor 
Schedule: Full-Time or Part-Time Contract
Fluent Language Skills Required: English &amp; German. Native-level fluency in English and German is required for this position.
Why This Role Exists
We believe the safest AI is the one that&rsquo;s already been attacked &mdash; by us. We are assembling a red team for this project - human data experts who probe AI models with adversarial inputs, surface vulnerabilities, and generate the red team data that makes AI safer for our customers.
This project involves reviewing AI outputs that touch on sensitive topics such as bias, misinformation, or harmful behaviors. All work is text-based, and participation in higher-sensitivity projects is optional and supported by clear guidelines and wellness resources. Before being exposed to any content, the topics will be clearly communicated.
What You&rsquo;ll Do
<ul><li>Red team conversational AI models and agents: jailbreaks, prompt injections, misuse cases, bias exploitation, multi-turn manipulation
</li><li>Generate high-quality human data: annotate failures, classify vulnerabilities, and flag systemic risks
</li><li>Apply structure: follow taxonomies, benchmarks, and playbooks to keep testing consistent
</li><li>Document reproducibly: produce reports, datasets, and attack cases customers can act on
</li></ul>Who You Are
<ul><li>You bring prior red teaming experience (AI adversarial work, cybersecurity, socio-technical probing)
</li><li>You&rsquo;re curious and adversarial: you instinctively push systems to breaking points
</li><li>You&rsquo;re structured: you use frameworks or benchmarks, not just random hacks
</li><li>You&rsquo;re communicative: you explain risks clearly to technical and non-technical stakeholders
</li><li>You&rsquo;re adaptable: thrive on moving across projects and customers
</li></ul>Nice-to-Have Specialties
<ul><li>Adversarial ML: jailbreak datasets, prompt injection, RLHF/DPO attacks, model extraction
</li><li>Cybersecurity: penetration testing, exploit development, reverse engineering
</li><li>Socio-technical risk: harassment/disinfo probing, abuse analysis, conversational AI testing
</li><li>Creative probing: psychology, acting, writing for unconventional adversarial thinking
</li></ul>What Success Looks Like
<ul><li>You uncover vulnerabilities automated tests miss
</li><li>You deliver reproducible artifacts that strengthen customer AI systems
</li><li>Evaluation coverage expands: more scenarios tested, fewer surprises in production
</li></ul><h2 class="jobSectionHeader">Contract and Payment Terms</h2>
<ul><li>You will be engaged as an independent contractor.</li>
<li>This is a fully remote role that can be completed on your own schedule.</li>
<li>Projects can be extended, shortened, or concluded early depending on needs and performance.</li>
<li>Payments are weekly on Stripe or Wise based on services rendered.</li></ul></div>

German & English Trainer - Remote

<div>Work Mode: Remote 
Engagement Type: Independent Contractor 
Schedule: Full-Time or Part-Time Contract 
Fluent Language Skills Required: English &amp; Korean. Native-level fluency in English and Korean is required for this position.
Role:
We believe the safest AI is the one that&rsquo;s already been attacked &mdash; by us. We are assembling a red team for this project - human data experts who probe AI models with adversarial inputs, surface vulnerabilities, and generate the red team data that makes AI safer for our customers.
This project involves reviewing AI outputs that touch on sensitive topics such as bias, misinformation, or harmful behaviors. All work is text-based, and participation in higher-sensitivity projects is optional and supported by clear guidelines and wellness resources. Before being exposed to any content, the topics will be clearly communicated.
What You&rsquo;ll Do
<ul><li>Red team conversational AI models and agents: jailbreaks, prompt injections, misuse cases, bias exploitation, multi-turn manipulation
</li><li>Generate high-quality human data: annotate failures, classify vulnerabilities, and flag systemic risks
</li><li>Apply structure: follow taxonomies, benchmarks, and playbooks to keep testing consistent
</li><li>Document reproducibly: produce reports, datasets, and attack cases customers can act on
</li></ul>Who You Are
<ul><li>You bring prior red teaming experience (AI adversarial work, cybersecurity, socio-technical probing)
</li><li>You&rsquo;re curious and adversarial: you instinctively push systems to breaking points
</li><li>You&rsquo;re structured: you use frameworks or benchmarks, not just random hacks
</li><li>You&rsquo;re communicative: you explain risks clearly to technical and non-technical stakeholders
</li><li>You&rsquo;re adaptable: thrive on moving across projects and customers
</li></ul>Nice-to-Have Specialties
<ul><li>Adversarial ML: jailbreak datasets, prompt injection, RLHF/DPO attacks, model extraction
</li><li>Cybersecurity: penetration testing, exploit development, reverse engineering
</li><li>Socio-technical risk: harassment/disinfo probing, abuse analysis, conversational AI testing
</li><li>Creative probing: psychology, acting, writing for unconventional adversarial thinking
</li></ul>What Success Looks Like
<ul><li>You uncover vulnerabilities automated tests miss
</li><li>You deliver reproducible artifacts that strengthen customer AI systems
</li><li>Evaluation coverage expands: more scenarios tested, fewer surprises in production
</li></ul><h2 class="jobSectionHeader">Contract and Payment Terms</h2>
<ul><li><ul><li>You will be engaged as an independent contractor.</li>
<li>This is a fully remote role that can be completed on your own schedule.</li>
<li>Projects can be extended, shortened, or concluded early depending on needs and performance.</li>
<li>Payments are weekly on Stripe or Wise based on services rendered.</li></ul></li></ul></div>

English & Korean Trainer - Remote

<div>Location: Remote
Type: Full-time or Part-time Contract Work
Fluent Language Skills Required: English &amp; Italian. Native-level fluency in English and Italian is required for this position.
Why This Role Exists
We believe the safest AI is the one that&rsquo;s already been attacked , by us. We are assembling a red team for this project - human data experts who probe AI models with adversarial inputs, surface vulnerabilities, and generate the red team data that makes AI safer for our customers.
This project involves reviewing AI outputs that touch on sensitive topics such as bias, misinformation, or harmful behaviors. All work is text-based, and participation in higher-sensitivity projects is optional and supported by clear guidelines and wellness resources. Before being exposed to any content, the topics will be clearly communicated.
What You&rsquo;ll Do
<ul><li>Red team conversational AI models and agents: jailbreaks, prompt injections, misuse cases, bias exploitation, multi-turn manipulation
</li><li>Generate high-quality human data: annotate failures, classify vulnerabilities, and flag systemic risks
</li><li>Apply structure: follow taxonomies, benchmarks, and playbooks to keep testing consistent
</li><li>Document reproducibly: produce reports, datasets, and attack cases customers can act on
</li></ul>Who You Are
<ul><li>You bring prior red teaming experience (AI adversarial work, cybersecurity, socio-technical probing)
</li><li>You&rsquo;re curious and adversarial: you instinctively push systems to breaking points
</li><li>You&rsquo;re structured: you use frameworks or benchmarks, not just random hacks
</li><li>You&rsquo;re communicative: you explain risks clearly to technical and non-technical stakeholders
</li><li>You&rsquo;re adaptable: thrive on moving across projects and customers
</li></ul>Nice-to-Have Specialties
<ul><li>Adversarial ML: jailbreak datasets, prompt injection, RLHF/DPO attacks, model extraction
</li><li>Cybersecurity: penetration testing, exploit development, reverse engineering
</li><li>Socio-technical risk: harassment/disinfo probing, abuse analysis, conversational AI testing
</li><li>Creative probing: psychology, acting, writing for unconventional adversarial thinking
</li></ul>What Success Looks Like
<ul><li>You uncover vulnerabilities automated tests miss
</li><li>You deliver reproducible artifacts that strengthen customer AI systems
</li><li>Evaluation coverage expands: more scenarios tested, fewer surprises in production</li></ul></div>

Italian Ai Trainer - Remote

Search 7200+ Adversarial ai tester jobs near you this March. New roles added regularly across leading industries. Apply today.

7200 adversarial ai tester jobs Near Me

Spanish Ai Trainer - Remote

German & English Trainer - Remote

English & Korean Trainer - Remote

English & Korean Trainer - Remote

Italian Ai Trainer - Remote

Italian Ai Trainer - Remote

German & English Trainer - Remote