Generative AI Content Evaluator

Assess generative AI content through careful comparisons, replay, and zoomed inspection. This remote freelance opportunity offers flexible hours for fluent English speakers.

Outlier is seeking an English-speaking evaluator to help train voice, image, and video models. The work involves judging whether outputs follow prompts and rating their quality and realism. You will document observations clearly while making decisions between imperfect options.

Key Responsibilities:

• Compare pairs of generated images and videos
• Judge which output better follows each prompt
• Replay and zoom into outputs to identify problems
• Detect distorted anatomy, flickering, motion, and text errors
• Identify audio that does not match video

Requirements:

• Fluent English and natural conversational speaking ability
• Patience to look and listen closely
• Ability to think on your feet
• Clear decision-making when both options have flaws
• Own ratings and explanations without AI writing tools

Use close visual and audio inspection, prompt comparison, and clear judgment to evaluate generative AI content for Outlier.