AI Safety & Policy Associate – Intermediate

Job Description

About the Role:

At Micro AI, we see AI safety as a living system, always evolving, always being tested, never finished. As an AI Safety & Policy Associate, you serve as one of the core guardians of our training data ecosystem. Your role is to help ensure that AI systems learn responsibly by carefully reviewing, labeling, and evaluating content across a wide range of sensitive and complex scenarios.

While policy designers define the rules, you are the one who brings those rules to life across thousands of real-world examples, including the ambiguous “gray zones” where context matters as much as correctness.

Your work helps shape how AI understands safety long before it ever interacts with a user.

What You’ll Do (Key Responsibilities):

Safety Annotation
Evaluate AI-generated responses using structured safety guidelines, identifying issues such as bias, misinformation, harmful framing, or restricted content.

Harm Classification
Label and categorize different types of harm, including hate speech, self-harm content, illegal activity, or unsafe guidance, helping models learn clear boundaries.

Fact-Checking & Verification
Cross-check AI outputs against reliable sources to detect inaccuracies and reduce hallucinated or unsupported claims.

Policy Application in Edge Cases
Apply detailed safety policies to complex or unclear scenarios where judgment is required to interpret intent, context, and potential risk.

Feedback & Improvement Loop
Report recurring issues, failure patterns, and edge-case challenges to governance teams to help refine and strengthen AI safety frameworks.

Qualification Path

Mandatory Qualifications:

* Education: Bachelor’s degree in any discipline (Humanities, Social Sciences, Communications, or Law preferred)
* Language Ability: Native-level proficiency in your primary language, including understanding of cultural nuance, slang, and context-dependent meaning
* English Proficiency: Minimum B1 (Intermediate) level for guideline interpretation and reporting
* Attention to Detail: Strong ability to detect subtle inconsistencies, even in repetitive or high-volume review tasks

Assessment

Applicants must complete a role-specific qualification exam followed by identity verification to assess suitability for the program.

Compensation

* Estimated range: $16 – $20 USD per hour
* Payment terms vary depending on project assignment and scope

Why Join the Micro AI Community

* Global Collaboration: Work alongside a large, diverse network of AI contributors worldwide
* Real-World Impact: Help shape safer AI systems used at scale across industries
* Advanced Learning Exposure: Gain experience in AI alignment, evaluation, and policy enforcement
* Cross-Domain Network: Collaborate with specialists across multiple languages and disciplines
* Flexible Work Style: Choose your schedule while contributing to meaningful AI safety work

Equal Opportunity

Micro AI is committed to equal opportunity employment. All applicants are assessed based on qualifications, skills, and performance alone, without regard to race, gender identity, religion, nationality, disability, or any protected status.