Evaluate AI-generated media remotely on your own schedule, using fluent English, careful observation, and explicit written explanations. Work flexibly across varied projects.

Outlier Experts support Scale AI’s training platform by reviewing AI-generated images, videos, audio, and voice conversations. You will score outputs for adherence, quality, realism, naturalness, pronunciation, and clarity. The work includes comparing media, roleplaying realistic conversations, identifying artifacts, and explaining evidence in English. Tasks vary across projects.

Key Responsibilities:

- Conduct live English conversations with AI voice models
- Compare generated images and videos against text prompts
- Inspect outputs for anatomy, motion, text, and audio problems
- Rate adherence, quality, realism, naturalness,



and clarity
- Write specific evidence-based explanations of evaluation decisions

Requirements:

- Fluent English with strong, clear written communication
- Strong attention to detail and patience during reviews
- Comfort with natural conversational speech and realistic situations
- Experience writing short, evidence-based explanations or reviews
- Comfort making clear decisions when both options have flaws

Bring fluent English, careful media review, and evidence-based judgment to Outlier’s AI evaluation projects remotely and independently.

#J-18808-Ljbffr

📌 AI Media Evaluation Specialist (Timmins)
🏢 SCALE AI
📍 Timmins