Anthropic's Fable AI just crushed the freelance test—and the competition

Anthropic's new AI model, Fable 5.2, has achieved a milestone on the Remote Labor Index, successfully completing over 41% of real freelance projects at professional quality. This benchmark measures how often AI agents deliver work good enough for a paying client to accept. Just a year ago, the top system managed only 2.5% of tasks, and Fable 5.2 now performs over 16 times better than that baseline.

For engineers and freelancers, this leap signals that AI agents are rapidly closing the gap on practical, complex work. The model shows particular strength in editing audio, designing architecture, and creating 3D models—areas where previous AI systems regularly failed. While it still struggles with industrial design and animation, Fable 5.2's ability to handle long, complex projects previously out of reach marks a significant step toward AI that can genuinely augment—and potentially disrupt—skilled creative labor.