Skip to content
OpenTrain AIFor AI Companies

Thai AI Safety Evaluation Expert

Use Thai fluency, cultural judgment, and structured reasoning to evaluate sensitive AI prompts and conversations. This remote, part-time contract pays $18-$22 per hour and starts immediately.

OpenTrain AI

Generative AI & RLHF

100% Remote Hourly · $18–$22/hr

$18–$22/hr

Compensation

Worldwide

Eligibility

Entry

Experience

Sep 4, 2026

Posted

Open worldwide

Interested in this role?

Create a free OpenTrain account and apply in minutes.

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. Create a free account to discover projects, build a professional profile, and apply for opportunities that match your skills.

About AI Safety Evaluation Work

AI safety evaluation is part of the human work behind advanced AI systems. Contributors review prompts and model conversations, identify risks or problematic behavior, and provide structured judgments that help improve how models respond to people.

  • Work remotely with a computer and an internet connection.
  • Use language expertise and careful reasoning to shape the behavior of cutting-edge AI.
  • Build experience in prompt evaluation, classification, adversarial review, and model safety.

The Role

OpenTrain is seeking a Thai AI Safety Evaluation Expert for remote, part-time contract work. You will use native or near-native Thai fluency, business-level written English, cultural judgment, and consistent evaluation practices when working with sensitive topics, written prompts, and AI model conversations.

The role is suitable for entry-level candidates who can apply written guidelines carefully. Prior AI or machine learning experience is not required.

  • Role focus: Thai AI safety evaluation
  • Work type: Part-time contractor
  • Schedule: Default commitment of 7 hours per week
  • Pay: $18-$22 per hour
  • Start: Immediately
  • Location: Worldwide and remote

What You'll Do

You will combine prompt creation, classification, adversarial review, and clear written reasoning. Your judgments will help strengthen model behavior across sensitive and potentially dual-use situations.

  • Write expert-level prompts in Thai across sensitive subject areas.
  • Classify prompts and conversations using structured guidelines.
  • Identify adversarial phrasing and escalation patterns.
  • Review written model interactions with close attention to language and cultural context.
  • Document the reasoning behind your judgments clearly and consistently.
  • Apply careful Thai language and cultural judgment to safety-related evaluations.

Required Qualifications

You must be able to understand and produce nuanced Thai writing while communicating your reasoning in business-level English. The work requires careful judgment around sensitive and dual-use information, along with the ability to follow written evaluation guidelines consistently.

  • Native or near-native fluency in Thai.
  • Business-level written English.
  • Strong written reasoning and attention to detail.
  • Sound judgment when reviewing sensitive and dual-use information.
  • Ability to write Thai prompts and classify prompts and conversations using structured guidelines.
  • Bachelor's degree completed or in progress.
  • No prior AI or machine learning experience required.

Helpful Background

Experience reviewing, grading, or red-teaming written or technical content can help you succeed. Familiarity with Thai cultural context and the ways meaning can change in sensitive situations is also valuable.

  • Trust and safety
  • Content moderation
  • Policy evaluation
  • Adversarial testing
  • Identifying escalation patterns in language-based systems
  • Reviewing or grading written or technical content

Why This Work Matters

Modern AI models depend on people who can evaluate examples, identify risks, and explain nuanced judgments. By reviewing Thai prompts and conversations, you will contribute language and cultural expertise to the development of safer AI systems while building experience in a fast-growing technology field.

  • Flexible part-time work that can fit around other commitments.
  • Remote access to work at the intersection of language and AI.
  • A chance to develop experience in safety evaluation and human feedback.

Apply Through OpenTrain

Create a free OpenTrain account to build your profile and apply for this opportunity. OpenTrain helps contributors discover AI training work, present relevant experience, and grow a durable career portfolio in the industry.

  • Review the role requirements.
  • Create or update your OpenTrain profile.
  • Apply with your Thai language and evaluation experience.
  • Prepare for remote, part-time contract work beginning immediately.

Ready to apply?

Create a free OpenTrain account and apply for this role in minutes.

Keep exploring

Similar jobs

View all AI training jobs

Thai Scientific AI Safety Evaluator

Use doctoral-level chemistry or biology expertise to write Thai scientific prompts and evaluate AI responses for accuracy, safety, and responsible handling of sensitive topics. This remote contractor role offers 7 hours per week at $24 to $28 per hour.

Generative AI & RLHF
Text
Remote · Worldwide
Thai, English
Part-time · Flexible
Entry level
Hourly · $24–$28/hr

Posted Sep 4, 2026

Vietnamese AI Safety Evaluator

Review sensitive AI conversations and create Vietnamese safety prompts using cultural judgment and careful reasoning. This remote contractor role offers $18-$22 per hour for about 7 hours weekly.

Generative AI & RLHF
Text
Remote · Worldwide
Vietnamese, English
Part-time · Flexible
Entry level
Hourly · $18–$22/hr

Posted Sep 4, 2026

AI Safety Red Team Expert English and Thai

Use expert adversarial testing to expose jailbreaks, prompt injections, bias, misinformation, and harmful behaviors in conversational AI. This remote contractor role offers $24-$35 per hour and about 40 hours per week.

Generative AI & RLHF
Text
Remote · Worldwide
Thai, English
Part-time · Flexible
Expert level
Hourly · $24–$35/hr

Posted Jul 30, 2026